{ "complete": true, "report": { "bucket": "casual_not_formal", "label": "Very informal/casual, not formal", "definition": { "id": "casual_not_formal", "category": "contrast", "head": "vn_S_CASU_reg", "pole": "high", "label": "Very informal/casual, not formal", "floor": 4.0, "opposite": "vn_S_FORM_reg", "canonical_dimension_name": "Casual Style" }, "recipe_status": "new_canonical_dimension_extension", "manual_provenance": [], "selected_wording": { "id": "canonical-extension-P09", "general": "Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.", "inline": [ "Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.", "Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.", "Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence." ], "recipe": "P09", "cfg": 1.0, "timing_syntax": "canonical", "canonical_labels": true, "lexical_proxy": false, "origin": "new_canonical_manual_extension_requires_new_evidence" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": -0.2181243760615638, "target_raw": 3.5528976917266846, "ce": 6.008912515640259, "genuineness": 4.347935724258423, "identity": 0.5706270158290863, "wer": 0.02245614035087719, "valid_fraction": 0.95, "identity_delta": -0.025375288724899293, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "manual_adapter": { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": -0.2877807582668326, "target_raw": 3.4764169633388518, "ce": 6.053788590431213, "genuineness": 4.402223420143128, "identity": 0.6116538271307945, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": 0.015651522576808928, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": -0.2181243760615638, "target_raw": 3.5528976917266846, "ce": 6.008912515640259, "genuineness": 4.347935724258423, "identity": 0.5706270158290863, "wer": 0.02245614035087719, "valid_fraction": 0.95, "identity_delta": -0.025375288724899293, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": -0.38688354742234615, "target_raw": 3.3699934482574463, "ce": 6.108593010902405, "genuineness": 4.416344499588012, "identity": 0.5620108991861343, "wer": 0.019298245614035085, "valid_fraction": 0.95, "identity_delta": -0.03399140536785126, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": -0.4187378116170968, "target_raw": 3.325480246543884, "ce": 6.077020907402039, "genuineness": 4.330676603317261, "identity": 0.5662543267011643, "wer": 0.023976608187134502, "valid_fraction": 0.95, "identity_delta": -0.02974797785282135, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": -0.6346130552051558, "target_raw": 3.072512185573578, "ce": 6.133215308189392, "genuineness": 4.325028121471405, "identity": 0.5822399973869323, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.013762307167053223, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.8083984428984947, "target_raw": 2.8749130249023436, "ce": 6.199735188484192, "genuineness": 4.3987716913223265, "identity": 0.5924243807792664, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.003577923774719238, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": -0.8606811551888741, "target_raw": 2.7879737496376036, "ce": 6.1504905939102175, "genuineness": 4.357977294921875, "identity": 0.5527782052755356, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.043224099278450015, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": -0.8671508532789013, "target_raw": 2.7852948904037476, "ce": 6.155256199836731, "genuineness": 4.3168693542480465, "identity": 0.5890215441584588, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.006980760395526886, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": -0.6863769599637451, "target_raw": 3.0307246506214143, "ce": 6.121698427200317, "genuineness": 4.337580227851868, "identity": 0.5448664359748363, "wer": 0.018421052631578946, "valid_fraction": 0.95, "identity_delta": -0.05113586857914924, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": -0.2877807582668326, "target_raw": 3.4764169633388518, "ce": 6.053788590431213, "genuineness": 4.402223420143128, "identity": 0.6116538271307945, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": 0.015651522576808928, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": -0.3246826285391991, "target_raw": 3.4306577920913695, "ce": 6.098466086387634, "genuineness": 4.323145413398743, "identity": 0.603341107070446, "wer": 0.016929824561403507, "valid_fraction": 0.95, "identity_delta": 0.007338802516460419, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": -0.3681213601684533, "target_raw": 3.4120768666267396, "ce": 6.069673919677735, "genuineness": 4.4031648397445675, "identity": 0.5862677961587905, "wer": 0.010526315789473684, "valid_fraction": 0.95, "identity_delta": -0.009734508395195008, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": -0.3800232026279843, "target_raw": 3.4206119537353517, "ce": 6.080198001861572, "genuineness": 4.325655746459961, "identity": 0.5832844406366349, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.01271786391735077, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": -0.383804335458671, "target_raw": 3.3666525959968565, "ce": 6.073248124122619, "genuineness": 4.286744260787964, "identity": 0.5796165764331818, "wer": 0.029239766081871343, "valid_fraction": 0.95, "identity_delta": -0.016385728120803834, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": -0.39752196914197163, "target_raw": 3.3874373555183412, "ce": 6.128052568435669, "genuineness": 4.364567172527313, "identity": 0.5605398327112198, "wer": 0.024871001031991743, "valid_fraction": 0.95, "identity_delta": -0.03546247184276581, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": -0.4259765594531783, "target_raw": 3.3184559822082518, "ce": 6.092310523986816, "genuineness": 4.3469942808151245, "identity": 0.548261896520853, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.04774040803313255, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": -0.44154662953919244, "target_raw": 3.3412576794624327, "ce": 6.15565333366394, "genuineness": 4.382140135765075, "identity": 0.5519146807491779, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.044087623804807664, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": -0.460653694564729, "target_raw": 3.3382750749588013, "ce": 6.131031060218811, "genuineness": 4.34260104894638, "identity": 0.5655099391937256, "wer": 0.003333333333333333, "valid_fraction": 1, "identity_delta": -0.03049236536026001, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": -0.6951904234120732, "target_raw": 2.963627851009369, "ce": 6.139172291755676, "genuineness": 4.336952650547028, "identity": 0.5749827563762665, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.021019548177719116, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": -0.7218932999370509, "target_raw": 2.9972853004932403, "ce": 6.1961610317230225, "genuineness": 4.316869246959686, "identity": 0.5885430194437504, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.007459285110235215, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": -0.7990234417347515, "target_raw": 2.8868122577667235, "ce": 6.151682019233704, "genuineness": 4.337580275535584, "identity": 0.5842470467090607, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": -0.011755257844924927, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": -0.8107177713237707, "target_raw": 2.8798347234725954, "ce": 6.171141600608825, "genuineness": 4.345111513137818, "identity": 0.5767675206065178, "wer": 0.058615196078431374, "valid_fraction": 0.9, "identity_delta": -0.019234783947467804, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.8299072223824336, "target_raw": 2.8187186658382415, "ce": 6.17709858417511, "genuineness": 4.3830815076828005, "identity": 0.5729297436773777, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.023072560876607896, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.875836168914681, "target_raw": 2.7930356085300447, "ce": 6.181467056274414, "genuineness": 4.322517728805542, "identity": 0.5600326254963874, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.03596967905759811, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": -0.16744995937283969, "target_raw": 3.587793302536011, "ce": 6.008316779136658, "genuineness": 4.279526817798614, "identity": 0.545006624609232, "wer": 0.0381765995872033, "valid_fraction": 0.95, "identity_delta": -0.050995679944753645, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": -0.16779480365742352, "target_raw": 3.6282180666923525, "ce": 6.040087461471558, "genuineness": 4.259129679203033, "identity": 0.48818167522549627, "wer": 0.03715793773649811, "valid_fraction": 1, "identity_delta": -0.1078206293284893, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": -0.20223084490978713, "target_raw": 3.5289979338645936, "ce": 6.019237923622131, "genuineness": 4.289568495750427, "identity": 0.5444387339055539, "wer": 0.019846491228070175, "valid_fraction": 0.95, "identity_delta": -0.05156357064843178, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": -0.6063232384131148, "target_raw": 3.113988256454468, "ce": 6.165581703186035, "genuineness": 4.389043807983398, "identity": 0.5420983508229256, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.05390395373106003, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": -0.6204528718360857, "target_raw": 3.056828272342682, "ce": 6.126861166954041, "genuineness": 4.339149212837219, "identity": 0.5311323598027229, "wer": 0.025964912280701753, "valid_fraction": 0.95, "identity_delta": -0.06486994475126266, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": -0.6606811640757736, "target_raw": 3.071717858314514, "ce": 6.175907182693481, "genuineness": 4.402537250518799, "identity": 0.5276057228446007, "wer": 0.011721491228070174, "valid_fraction": 1, "identity_delta": -0.06839658170938492, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.7459381093308596, "target_raw": 2.931816041469574, "ce": 6.228328847885132, "genuineness": 4.389985156059265, "identity": 0.5146423742175102, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.08135993033647537, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.8122558698173169, "target_raw": 2.7942348659038543, "ce": 6.15485908985138, "genuineness": 4.326910984516144, "identity": 0.5448040351271629, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.051198269426822665, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": -0.8396972674857234, "target_raw": 2.8265061020851134, "ce": 6.186629843711853, "genuineness": 4.338521575927734, "identity": 0.5454360276460648, "wer": 0.005263157894736842, "valid_fraction": 1, "identity_delta": -0.050566276907920836, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -0.923974616831138, "target_raw": 2.6237204909324645, "ce": 6.130832505226135, "genuineness": 4.397516417503357, "identity": 0.5960023045539856, "wer": 0.002631578947368421, "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": -0.9853088416437109, "target_raw": 2.582182115316391, "ce": 6.15485908985138, "genuineness": 4.402223420143128, "identity": 0.5873977646231652, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.008604539930820465, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -1.0339721721986088, "target_raw": 2.5163469076156617, "ce": 6.148107814788818, "genuineness": 4.343228685855865, "identity": 0.5509783059358597, "wer": 0.0027777777777777775, "valid_fraction": 1, "identity_delta": -0.045023998618125914, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -0.9585144121600098, "target_raw": 2.628782331943512, "ce": 6.171141576766968, "genuineness": 4.400968182086944, "identity": 0.5850818671286107, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.010920437425374985, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.9862597692964027, "target_raw": 2.703663535118103, "ce": 6.233412141799927, "genuineness": 4.391993470191956, "identity": 0.6015608149766922, "wer": 0.0076364825106243865, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": -0.3753100617203283, "target_raw": 3.495331163406372, "ce": 6.146042718887329, "genuineness": 4.373039875030518, "identity": 0.5753106416016817, "wer": 0.022835309617918312, "valid_fraction": 1, "identity_delta": -0.02625017337501049, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": -0.5323107915282019, "target_raw": 3.3067342925071714, "ce": 6.178607683181763, "genuineness": 4.368646602630616, "identity": 0.5903049494326115, "wer": 0.01594635798983625, "valid_fraction": 0.98, "identity_delta": -0.011255865544080734, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": -0.4420214990178078, "target_raw": 3.399321026802063, "ce": 6.125868330001831, "genuineness": 4.390863819122314, "identity": 0.5562701150774956, "wer": 0.03873708480687885, "valid_fraction": 0.96, "identity_delta": -0.045290699899196624, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": -0.09028929251039414, "ci95_z": [ -0.193347154719618, 0.055709246331492254 ], "language_deltas": { "en": 0.18636477493289066, "de": -0.2122619588972646, "fr": -0.1723144328837407, "nl": -0.18494871145991, "es": -0.06828613424394608 }, "languages_improved": 1, "clear_gain": false, "delta_ce": 0.05273935317993164, "delta_genuineness": -0.022217216491699218, "delta_wer": -0.022790726817042608, "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": -0.15700072980787363, "ci95_z": [ -0.2092333811068628, -0.0916577055208151 ], "language_deltas": { "en": -0.18034667024236997, "de": -0.1392151321112478, "fr": -0.23530271087668778, "nl": -0.03253172903977852, "es": -0.19760740676928415 }, "languages_improved": 0, "clear_gain": false, "delta_ce": 0.03256496429443359, "delta_genuineness": -0.004393272399902344, "delta_wer": -0.006888951628082063, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.6109497075760744, "ci95_z": [ 0.4920678677417422, 0.7295703314302097 ], "language_deltas": { "en": 0.7808349860267411, "de": 0.5783203785265086, "fr": 0.7270263205443168, "nl": 0.3849975465459619, "es": 0.5835693062368432 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.08736942291259765, "delta_genuineness": -0.018953595161437988, "delta_wer": 0.015198827107293925, "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": 1793036707.5059724, "example": { "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.", "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." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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.", "language": "fr", "seed": 1701, "scale": 0.0, "cfg": 5.0, "frames": 128, "reference": true } }, "rows": [ { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "00902d6791756e9204ff5ae1a34891f6deec29361a102eba58674f4cd26e5981", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9abeefa1711bccb0382032c9bc83188f30a32d4e5af33d7b6142ba21a895a385", "generation_seconds_batch": 21.779498178046197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7984442710876465, "target_z": 0.034179675390348316, "ce": 5.998785495758057, "genuineness": 4.469690799713135, "identity": 0.5792657136917114, "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/casual_not_formal/00902d6791756e9204ff5ae1a34891f6deec29361a102eba58674f4cd26e5981.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "00bc82abe723591d1a6b6026266b76a772a3071cd76ca510f3973865a7e08729", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "558dee748e5555513b86ce59da64fe059542059af9a934d8eb2633e1812205c5", "generation_seconds_batch": 23.022883794968948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.079416275024414, "target_z": 0.17602547624955472, "ce": 5.8875885009765625, "genuineness": 4.3127899169921875, "identity": 0.38639017939567566, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar.", "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/casual_not_formal/00bc82abe723591d1a6b6026266b76a772a3071cd76ca510f3973865a7e08729.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "00cb8b911fba67d5c6f4f83fb6766c6785b8c41e6a8e16000e8281b1c4b4144b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c18b52096a7570ad3bb23b5ad2e78fbf4faa7259ecfb95632f301fb0e7081a5f", "generation_seconds_batch": 25.191665918100625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.619110345840454, "target_z": -0.757812491587343, "ce": 6.229123115539551, "genuineness": 4.268857479095459, "identity": 0.6909728050231934, "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/casual_not_formal/00cb8b911fba67d5c6f4f83fb6766c6785b8c41e6a8e16000e8281b1c4b4144b.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": "00e08447732564853b6e94b55a93ede2140ce00197e0303e6eee26cec56b7630", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "5d634629b2175b048c4587136a762cf62e25908df98431c2454c10358a69ea4a", "generation_seconds_batch": 17.74569149094168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.274503469467163, "target_z": -0.49902347317235374, "ce": 5.911416053771973, "genuineness": 4.645420074462891, "identity": 0.8005795478820801, "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/casual_not_formal/00e08447732564853b6e94b55a93ede2140ce00197e0303e6eee26cec56b7630.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "00f447e8a389debd6ef6f4a7bdbf8220c99e1bebadfd5f912f6597089b35d844", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de19e9b7032ee5c50cfb07dc6a764b69a76b9ee87324df0aa6dab12aca3a5548", "generation_seconds_batch": 23.341999776894227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.617685794830322, "target_z": 0.40820310652032826, "ce": 6.062326908111572, "genuineness": 4.444586753845215, "identity": 0.656557559967041, "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/casual_not_formal/00f447e8a389debd6ef6f4a7bdbf8220c99e1bebadfd5f912f6597089b35d844.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "00fec27e56427cf6f90554d7417bd4b0df74da96f793f23048bbddaf5a49c543", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ce6d6c1ce3d13cc43927900696138eb778634cf38726d630d3ec71cbc08014e", "generation_seconds_batch": 24.573775270953774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -1.007812565249112, "ce": 6.229123115539551, "genuineness": 4.457139015197754, "identity": 0.7585890293121338, "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/casual_not_formal/00fec27e56427cf6f90554d7417bd4b0df74da96f793f23048bbddaf5a49c543.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "01074caf82d5a94f19ddbb96d8d5a991b920c16de1a7e4b4a493374033ff073a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b84b897b1c81ca5fe4a262192326fbd0fc59dd6d22fb7d2497dcca5831eab553", "generation_seconds_batch": 25.824389254907146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4347031116485596, "target_z": -1.1171874502139767, "ce": 6.23706579208374, "genuineness": 4.557555198669434, "identity": 0.683301568031311, "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/casual_not_formal/01074caf82d5a94f19ddbb96d8d5a991b920c16de1a7e4b4a493374033ff073a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "014e5f8c3d8558a7661c271c9846406f0af07f5200e858e8246a116b911910e9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae90439a175769dc76a00d59018be7d12d1c033a46f345c17f57e0156f4c5fe0", "generation_seconds_batch": 26.462575230980292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4701247215270996, "target_z": -0.31884765127011666, "ce": 6.022613525390625, "genuineness": 4.256305694580078, "identity": 0.4570716619491577, "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/casual_not_formal/014e5f8c3d8558a7661c271c9846406f0af07f5200e858e8246a116b911910e9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "01801e37c0aba084211f0603233dd605412636d07eb4e444ae24b368771c9feb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "02e75f8a646fb1ffdf16ac1ed813db5c0ee0f98a979878881a11c90c5dd144bf", "generation_seconds_batch": 21.247167141991667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0260167121887207, "target_z": -1.464843857309995, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.5105627775192261, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/01801e37c0aba084211f0603233dd605412636d07eb4e444ae24b368771c9feb.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "afc345b321ba4578be6f5e395fc06bbf5e86c399a64ce1ebac91d34934dc34e3", "generation_seconds_batch": 17.182745361002162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9430689811706543, "target_z": -0.832031219573251, "ce": 6.022613525390625, "genuineness": 4.68307638168335, "identity": 0.8181737661361694, "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/casual_not_formal/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "022f6670e31941a47c4f3e6b58eb1fc96b7a8583db85202f933e35d45725bb5f", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b6657405c6468de76a737387f2d64e98787372cdc2e5c0151d1e823e26ae5c5b", "generation_seconds_batch": 19.10751733998768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5892064571380615, "target_z": -1.103515600163014, "ce": 6.197352409362793, "genuineness": 4.394378662109375, "identity": 0.6545660495758057, "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/casual_not_formal/022f6670e31941a47c4f3e6b58eb1fc96b7a8583db85202f933e35d45725bb5f.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": "0258ab154705c0bdbd390a2ca5cb9aa352d8acb9398bed5edde9a9170176dca1", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "52da728555d993fd7c876cfb4050947e83bf12ac7e2b96de85cfa9ff4f569667", "generation_seconds_batch": 18.040472951019183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.154175758361816, "target_z": 0.15039059131415733, "ce": 6.070269584655762, "genuineness": 4.582659721374512, "identity": 0.612908661365509, "hypothesis": "La route tourna à gauche après le pont, continue jusqu'à la petite barrière en bois. J't'attendrai de l'autre côté.", "wer": 0.15, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/0258ab154705c0bdbd390a2ca5cb9aa352d8acb9398bed5edde9a9170176dca1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "026c21fe6971f8ca10ed0e58938cb181fa488116ef483a60f2ffbf8aa71eeed6", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "30be8eff45c2da375a20f1b82acd8a57bf944a405592261de73236845c23ba6d", "generation_seconds_batch": 17.143384692026302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.8671874897521006, "ce": 6.078212261199951, "genuineness": 4.419482707977295, "identity": 0.6473114490509033, "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/casual_not_formal/026c21fe6971f8ca10ed0e58938cb181fa488116ef483a60f2ffbf8aa71eeed6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "02742472ab75569c3aad89b806ab8ad5a030fb96ec400db0b718614700817fdf", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be29f1fc4cc3dc1b4baf00ba13ded9c914e90f31ab2b0adbb0003813dff72042", "generation_seconds_batch": 23.805179105023853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8433001041412354, "target_z": 0.117431660508515, "ce": 5.990842819213867, "genuineness": 4.582659721374512, "identity": 0.7309707403182983, "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/casual_not_formal/02742472ab75569c3aad89b806ab8ad5a030fb96ec400db0b718614700817fdf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "02915131a2403a872c730a79f37995edeb7d513baf0b22650b26daa153fd8359", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44c7ce0a7b98e260b1f111eea9afb86e2f8ec59f5915fd8ed1ae42e75d0fd1bf", "generation_seconds_batch": 27.408408271032386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0178287029266357, "target_z": -0.6093749360514833, "ce": 6.022613525390625, "genuineness": 4.268857479095459, "identity": 0.5668039321899414, "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/casual_not_formal/02915131a2403a872c730a79f37995edeb7d513baf0b22650b26daa153fd8359.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "0349cae9a6a2c83e5af3a7df202b9ccc0336962f288c4c350f2f02abfccbebc4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fe09313fed549ad6855c71fc1dac45410f3d2256c09b5e5828a19cddc1a0b7a1", "generation_seconds_batch": 27.497890384984203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2154078483581543, "target_z": -1.3125000083846463, "ce": 6.284721851348877, "genuineness": 4.557555198669434, "identity": 0.48838770389556885, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie standen neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/0349cae9a6a2c83e5af3a7df202b9ccc0336962f288c4c350f2f02abfccbebc4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "035416aa9ff803d4b3523cafe8c6bfd7d3ff60be023392c5d6d4719a6d7f8420", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "978c03e51d14bf7c4a483bd0b6d024560e7ed2d421d5d62dcbecacb5290a6a02", "generation_seconds_batch": 24.057310626027174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.567490816116333, "target_z": -1.957031311503394, "ce": 6.3641486167907715, "genuineness": 4.042920112609863, "identity": 0.5642202496528625, "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/casual_not_formal/035416aa9ff803d4b3523cafe8c6bfd7d3ff60be023392c5d6d4719a6d7f8420.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": "0369784f9c781de0ee0d98062ac92bc6cd822bc539481c36c26160baa714a312", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a2373c421afcb2132f4373b82c8fc738d3331bbd35af5521441cbbaf1bd7058", "generation_seconds_batch": 17.461518369033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0726523399353027, "target_z": -0.7578124920443536, "ce": 5.990842819213867, "genuineness": 4.632867813110352, "identity": 0.7109335660934448, "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/casual_not_formal/0369784f9c781de0ee0d98062ac92bc6cd822bc539481c36c26160baa714a312.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "044a3f16ebefbcd8021012cefb1cca9dc329002ff2c1ac96261be74a973bff37", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39dcf13cf3bb354873e26a6302c4e1c27c6a95fbd4178e468b6fb0fedd50ca46", "generation_seconds_batch": 17.545386334997602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.8046874888060671, "ce": 6.260893821716309, "genuineness": 4.325342178344727, "identity": 0.7952409982681274, "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/casual_not_formal/044a3f16ebefbcd8021012cefb1cca9dc329002ff2c1ac96261be74a973bff37.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": "0482d5921c47aa80ec041520eaa8d94995189ca864bc13dee67d490fdb7f2e13", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f837a81a87442ddcf9fc7ced982eb1cd48e42e2017845b4d5e153f80c5daf1ed", "generation_seconds_batch": 15.840128924930468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1175081729888916, "target_z": -0.4257812858460328, "ce": 6.38003396987915, "genuineness": 4.400654315948486, "identity": 0.17344240844249725, "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/casual_not_formal/0482d5921c47aa80ec041520eaa8d94995189ca864bc13dee67d490fdb7f2e13.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "0486a2b09f07c24d3d89324b7680e4065c02937bd6905ef538e78f31ba4ad774", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5ac3c3638a4bfa46eb2274a132787cf42b075bac344ec54f3b576fb8e1aee77", "generation_seconds_batch": 24.845025430084206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.335937521837213, "ce": 6.1894097328186035, "genuineness": 4.337893962860107, "identity": 0.7918381690979004, "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/casual_not_formal/0486a2b09f07c24d3d89324b7680e4065c02937bd6905ef538e78f31ba4ad774.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": "04c84648d65bf324bc07ad0071d77a198e8d33ddd16a5e39937c3d4163d9f61c", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "84cff5edfd39b09f5dc155ded7ae04069473bc8e9b410283fc5c02fbdc0c158c", "generation_seconds_batch": 16.726159596000798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.306899309158325, "target_z": -0.4375000274357783, "ce": 5.895530700683594, "genuineness": 4.545003414154053, "identity": 0.8073941469192505, "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/casual_not_formal/04c84648d65bf324bc07ad0071d77a198e8d33ddd16a5e39937c3d4163d9f61c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "0558292ec839f8ae102b54e001673f8960f44b40551cc160e58b20f24f1ce457", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c89b96147268a4b09999ff8dcb674cb27507d7e28f54712614bb58859c918f7", "generation_seconds_batch": 25.368043602094986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.228935241699219, "target_z": 0.12353504568271026, "ce": 6.078212261199951, "genuineness": 4.2374773025512695, "identity": 0.5360825657844543, "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/casual_not_formal/0558292ec839f8ae102b54e001673f8960f44b40551cc160e58b20f24f1ce457.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "057fd84dad1790f180a015446431f14cea23cb1cfb67fe38e118a124ba60e519", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f4b07d7fe4ba32e7b94ab89decceb49f9a69cf1859c3a5f820d75352300ac0d7", "generation_seconds_batch": 22.52777690498624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9286508560180664, "target_z": 0.08154301996948181, "ce": 5.744620323181152, "genuineness": 4.344170093536377, "identity": 0.669033944606781, "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/casual_not_formal/057fd84dad1790f180a015446431f14cea23cb1cfb67fe38e118a124ba60e519.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "8db51932707269e5693624245427ec27f38204f8bc3ed525cf6a0afc8d6beadc", "generation_seconds_batch": 17.828491030959412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.022812604904175, "target_z": -0.6328125294294908, "ce": 6.308549880981445, "genuineness": 4.444586753845215, "identity": 0.686214804649353, "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/casual_not_formal/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "05b4c8b7764ef6cfb265a832c2829f7a65b68fe357c520ad839b7bbcb71f5d27", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6776965a5fdd72b51141dcc062834c8765056fc16e0f8312d1016f60e6d64f8a", "generation_seconds_batch": 17.309360736981034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.406249857064073, "ce": 6.403861999511719, "genuineness": 4.356721878051758, "identity": 0.5780056715011597, "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/casual_not_formal/05b4c8b7764ef6cfb265a832c2829f7a65b68fe357c520ad839b7bbcb71f5d27.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": "060fab56767bb633e1a5cc8c45ed9dc25886782535a7d8c11aadb10965b4d85a", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "1bd6d03d38cb627893bbe110590cbef74d6da7993ebf7e15ed4548b64f2bc965", "generation_seconds_batch": 18.145177373080514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.142428159713745, "target_z": -0.7382811613257487, "ce": 6.205295085906982, "genuineness": 4.268857479095459, "identity": 0.7320266962051392, "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/casual_not_formal/060fab56767bb633e1a5cc8c45ed9dc25886782535a7d8c11aadb10965b4d85a.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": "0619aa8f4179f0a3c3afb617a6e98a818d6b1c23b723073e6a20f91499bcba37", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67280d5b682ba2cddd99aff8dba2e222c43850604c16da4633ae30aeb068de5d", "generation_seconds_batch": 24.029069230891764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.914944887161255, "target_z": -0.08007809179917873, "ce": 6.229123115539551, "genuineness": 4.3504462242126465, "identity": 0.6075030565261841, "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/casual_not_formal/0619aa8f4179f0a3c3afb617a6e98a818d6b1c23b723073e6a20f91499bcba37.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "0623179998e10fbc716db42583ed21766c108d50a5251465c67ad59d1c143c42", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ad56bd127a614d4ebb268b317843d553afd455ff1f0aff431f75141e5b94c8d", "generation_seconds_batch": 16.794543407973833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0028767585754395, "target_z": -0.6054687330258175, "ce": 6.141753673553467, "genuineness": 4.457139015197754, "identity": 0.8256726861000061, "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/casual_not_formal/0623179998e10fbc716db42583ed21766c108d50a5251465c67ad59d1c143c42.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "0637f150b12375388fe8168837c9f6b126f3034a34b83de7e119e53c0de08e28", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bb6731b8510dc3d5157bea5c08229309fa342ab6bb68b5ed84c0382ec0172e6", "generation_seconds_batch": 19.731660658027977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.3203125930762694, "ce": 5.71682071685791, "genuineness": 4.18099308013916, "identity": 0.753153920173645, "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/casual_not_formal/0637f150b12375388fe8168837c9f6b126f3034a34b83de7e119e53c0de08e28.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "065ab87462cf251c08b218fbfcaeac99c392fe67c464a248f651409729f8996a", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "357c4dcc874305317740bfc986bccc16567272697aa1686a5203b7cdfa4f125b", "generation_seconds_batch": 16.597163792001083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0658886432647705, "target_z": -1.4492188178935241, "ce": 5.919358730316162, "genuineness": 4.15588903427124, "identity": 0.525105893611908, "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/casual_not_formal/065ab87462cf251c08b218fbfcaeac99c392fe67c464a248f651409729f8996a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "06e78bdd92a081cfd1dea4b0857522dcf701273f8ccf7bc268ff7514bdc30ac5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b91dc216ffa01af6ecd68ebc1f0da3a5a4dda17b06c01c1fd701f737ad8e020", "generation_seconds_batch": 17.162246962077916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9622926712036133, "target_z": 1.706182259364475e-08, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.8140219449996948, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu tiens, tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/06e78bdd92a081cfd1dea4b0857522dcf701273f8ccf7bc268ff7514bdc30ac5.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": "07a02c02434ef207f5c739412689b67c46306ef12684fff6dac95d79d2151835", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "0cfc79141a83a0cbc1b6e39ecf1bff5f26e7e7bb65eaef6f53cbb4d207fbd3be", "generation_seconds_batch": 16.803563880966976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6102991104125977, "target_z": -0.34399413651871735, "ce": 6.141753673553467, "genuineness": 4.645420074462891, "identity": 0.7531256079673767, "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/casual_not_formal/07a02c02434ef207f5c739412689b67c46306ef12684fff6dac95d79d2151835.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "07f381b897bf5d376a5c27c801c5ccda6eb403be0f1493be52f97ed9ca113ddb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dd0ff4a86bba22f8744820b55299e6f798d76dd0f99f2de3287f1b6d0e49c96", "generation_seconds_batch": 24.141512085916474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2271556854248047, "target_z": -0.6386718929170141, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.8570705056190491, "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/casual_not_formal/07f381b897bf5d376a5c27c801c5ccda6eb403be0f1493be52f97ed9ca113ddb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "0866ee39f55d7e4dc9fd72d707d16f05f6e7b24ae4939a4f6dbfd9b02edcca15", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0940aeb641e686ff78061b69b539750f2a937952202fc3f3f373e09ebb041660", "generation_seconds_batch": 25.484788287081756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.079416275024414, "target_z": 0.05078131455439572, "ce": 6.229123115539551, "genuineness": 4.344170093536377, "identity": 0.35934978723526, "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/casual_not_formal/0866ee39f55d7e4dc9fd72d707d16f05f6e7b24ae4939a4f6dbfd9b02edcca15.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "0883ab9ccb65d8493f682e2bc7f358319454fb215206ad06861feca0b1d97ac4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8052231d2e1fafd385843c7f83cd10b84533ddb934122cf0502fd7844fe94433", "generation_seconds_batch": 21.45494276296813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8352010250091553, "target_z": -0.02392585008815857, "ce": 6.070269584655762, "genuineness": 4.337893962860107, "identity": 0.5644588470458984, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaarbort.", "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/casual_not_formal/0883ab9ccb65d8493f682e2bc7f358319454fb215206ad06861feca0b1d97ac4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "08964457ef947ee937cbdfa6bf0e6d9f00ec4becb4f28f5f1355876725b8fd53", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bc77e5c6ae94a392c95c70749d7b65fb8ff93429deaf46893777e6ea5456748", "generation_seconds_batch": 21.516050491016358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6620078086853027, "target_z": -0.34326171810570566, "ce": 6.181467056274414, "genuineness": 4.532451152801514, "identity": 0.686507523059845, "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/casual_not_formal/08964457ef947ee937cbdfa6bf0e6d9f00ec4becb4f28f5f1355876725b8fd53.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "08a43c16d96b45e33248d8a01e88cb2d3a2c96d0140fa2c41de7ad5ee229ba79", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58149b8e908339992d96f627b0fb2135cf3075dff4c9c03b5b75f3a4a696dfaf", "generation_seconds_batch": 17.12671903090086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5199644565582275, "target_z": -0.2963867181789386, "ce": 6.078212261199951, "genuineness": 4.218649387359619, "identity": 0.6902990341186523, "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/casual_not_formal/08a43c16d96b45e33248d8a01e88cb2d3a2c96d0140fa2c41de7ad5ee229ba79.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "08b6bf651a14722ef9d248bda6de0f07a22d2472a031ba40a2b35f93e24ce0de", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df854014c92519da6943b12c8c5c4cc661665c8f9f537462a0342fd9b5a3d6a6", "generation_seconds_batch": 16.542153516900726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5959701538085938, "target_z": -0.26562502538843863, "ce": 6.372091293334961, "genuineness": 4.1998209953308105, "identity": 0.24579258263111115, "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/casual_not_formal/08b6bf651a14722ef9d248bda6de0f07a22d2472a031ba40a2b35f93e24ce0de.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "093c92b2f079c4a143b741d855fc9020709728253cbac5f5dbc517a674fe0ac7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dedbf7cb3486b8e6109901eea252c5a3f5f9b616c1e873b29f34530b122ec5a3", "generation_seconds_batch": 22.68779147497844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.064464092254639, "target_z": 0.03515620278075263, "ce": 6.086154937744141, "genuineness": 4.507347106933594, "identity": 0.5780684947967529, "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/casual_not_formal/093c92b2f079c4a143b741d855fc9020709728253cbac5f5dbc517a674fe0ac7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "099e08a0b9e925054225b15d57f185b5f0d83ed4d4a3cccb817de507d93ef3b6", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c5e9f1f94a79feec9f9ce28b9943ca3a14b10e3f257620bd9bfa707868117d4", "generation_seconds_batch": 18.46596200799104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9363051652908325, "target_z": -1.5781249646782085, "ce": 6.173524379730225, "genuineness": 4.13078498840332, "identity": 0.7651385068893433, "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/casual_not_formal/099e08a0b9e925054225b15d57f185b5f0d83ed4d4a3cccb817de507d93ef3b6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "09c040ddf60e37837e073b41bfbe6a55e742944fdaada66defe0a117ead114e3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc08d6d653d0317a6723cc2f9623bee112c3e7218dc469e3b1cccc1477520cd6", "generation_seconds_batch": 21.779498178046197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.533670425415039, "target_z": -0.19775388546667244, "ce": 6.117925643920898, "genuineness": 4.369274139404297, "identity": 0.45680901408195496, "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/casual_not_formal/09c040ddf60e37837e073b41bfbe6a55e742944fdaada66defe0a117ead114e3.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": "09ce03a4cbf834ce2562272fc7dc104adda34f854dae1c4dc0a59d40196df5b3", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93a8ff8fb14703b61fcc1404ee32d511432000e9a3022027788feaf1b6aff0c1", "generation_seconds_batch": 22.750562725006603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6408259868621826, "target_z": -0.2695312286750239, "ce": 6.316492557525635, "genuineness": 4.105680465698242, "identity": 0.38696807622909546, "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/casual_not_formal/09ce03a4cbf834ce2562272fc7dc104adda34f854dae1c4dc0a59d40196df5b3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "09dc5decad50b1c1130d9786ce910e1f73ee481cde2fbad157c97dc7fa3b30d0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a86fabd202735f34fcd0d7ed4b49eb590993bf01590b92544d410602e0509a41", "generation_seconds_batch": 18.756172602064908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6003310680389404, "target_z": 0.045165979245721924, "ce": 6.1020402908325195, "genuineness": 3.9676074981689453, "identity": 0.4208105206489563, "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/casual_not_formal/09dc5decad50b1c1130d9786ce910e1f73ee481cde2fbad157c97dc7fa3b30d0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "09ee56ca54658be36b22b49fd999102688be9438fe65dd8b434bf445a0f43525", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b735be461f642280b610adb883b709b419adbed047b4d2fb1e3bdd3c367ef363", "generation_seconds_batch": 26.07389770296868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7536778450012207, "target_z": -0.7890625001504674, "ce": 6.149696350097656, "genuineness": 4.369274139404297, "identity": 0.6476132869720459, "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/casual_not_formal/09ee56ca54658be36b22b49fd999102688be9438fe65dd8b434bf445a0f43525.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "0aa66af362ef67c78c943572c395267bbc3240c4b91604a4eff77e488318a9a2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "864d6e043a9ced1983b8ecda0b4634b181bc10e1102d5849b9c0647793db1a03", "generation_seconds_batch": 16.408794441027567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9530367851257324, "target_z": -0.7929688195086657, "ce": 6.141753673553467, "genuineness": 4.545003414154053, "identity": 0.5268236398696899, "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/casual_not_formal/0aa66af362ef67c78c943572c395267bbc3240c4b91604a4eff77e488318a9a2.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": "0b1bce28736a0f747008946426bca2b013b0db4b102db1be4c551aad1c8ca7c0", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 127, "seconds_generated": 10.16, "mp3_sha256": "0d67ec8ca1f745ed25ec0696781d80ad0caa4c9b96ae7480dc9efb2bc01531a2", "generation_seconds_batch": 16.54173002706375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9498326778411865, "target_z": 0.019287095753055816, "ce": 6.23706579208374, "genuineness": 4.331617832183838, "identity": 0.7463799118995667, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/0b1bce28736a0f747008946426bca2b013b0db4b102db1be4c551aad1c8ca7c0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "0b6019dde5b3de2e6d7e60cf116f8caaaf03446d434cf97835770bf6f4eb7af4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "615ee2ba3df9f0412429b802e274a6720412c126d35eec727cef0353589b43f9", "generation_seconds_batch": 26.45920089504216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8109042644500732, "target_z": -0.04125974601180857, "ce": 6.006728172302246, "genuineness": 4.356721878051758, "identity": 0.6120518445968628, "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/casual_not_formal/0b6019dde5b3de2e6d7e60cf116f8caaaf03446d434cf97835770bf6f4eb7af4.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": "0bcc3baeab9f8fb615aaf65187f0f63f1605333e84e25cffbc3c92e0ba5c22fa", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "1b7a5cb0e7c90ac8a6f76f35a6e9a1640490054960e5daec9b2626c3cacff4c2", "generation_seconds_batch": 16.766787564964034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9460947513580322, "target_z": 0.07812504099859588, "ce": 6.173524379730225, "genuineness": 4.494794845581055, "identity": 0.5889071822166443, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het na de lunch.", "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/casual_not_formal/0bcc3baeab9f8fb615aaf65187f0f63f1605333e84e25cffbc3c92e0ba5c22fa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "0bebc7ea299f837f04fccdd896d4f5928b96bc84445ad972774182140ca1a549", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50c658731f5455fddf91b3d36b3a3c09bacec92acf1f2a0758ab61adae4b1d91", "generation_seconds_batch": 16.814341040095314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.8085937670590182, "ce": 6.197352409362793, "genuineness": 4.507347106933594, "identity": 0.6397577524185181, "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/casual_not_formal/0bebc7ea299f837f04fccdd896d4f5928b96bc84445ad972774182140ca1a549.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "fdcc20974e4611e5a3addbb44e475e4566f8916d474e3e4bbeda4fc7463efeab", "generation_seconds_batch": 16.991411286988296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.5234376252830346, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.7467135190963745, "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/casual_not_formal/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "0ce3b3f263df51f456eafa268ebb0eef2c883cea50d842497baf7bd9643e0d9c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a34900854508fd5a737712c4c6677a2d2995f22d709c57055decb786957f14ff", "generation_seconds_batch": 22.75484538602177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5393667221069336, "target_z": -1.0839842684007315, "ce": 6.149696350097656, "genuineness": 4.243753433227539, "identity": 0.7484531402587891, "hypothesis": "He encontrado que le encontrado. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.17647058823529413, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/0ce3b3f263df51f456eafa268ebb0eef2c883cea50d842497baf7bd9643e0d9c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "0ceddc64b0462fb0faefbebccc9abc1cff5c6f03af0e25e2b28efa08d2539b00", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f2170d93968bc5e43cd493eec5be1d115ce4c865d7def77b2446a596b372ca2", "generation_seconds_batch": 21.083252437063493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.037764549255371, "target_z": -0.4863281328548005, "ce": 6.308549880981445, "genuineness": 3.9299514293670654, "identity": 0.40978503227233887, "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/casual_not_formal/0ceddc64b0462fb0faefbebccc9abc1cff5c6f03af0e25e2b28efa08d2539b00.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": "0d21dddf6f172506c2a73dd342706b318bdbd57d91125704b2f064f710bcfb16", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 128, "seconds_generated": 10.24, "mp3_sha256": "efe4bdae1f94bec5a8c787bd74b9fa90312aca796237449c8af2c755f809bf30", "generation_seconds_batch": 15.970739534008317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4863226413726807, "target_z": -0.478515647074119, "ce": 6.09409761428833, "genuineness": 4.469690799713135, "identity": 0.6836404204368591, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Mel me wanneer je even tijd hebt.", "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/casual_not_formal/0d21dddf6f172506c2a73dd342706b318bdbd57d91125704b2f064f710bcfb16.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "0d5b17b7a6aa84b43f9bcbfedcd9e45c4702cc6679b0b987c7b993ed4d11b36c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d0b217980f0f9aaa88dfd1326b11e9771b45f49cd83f53dd07e7076c8cbf49ac", "generation_seconds_batch": 22.362677935045213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.388422966003418, "target_z": 0.5546875608374919, "ce": 6.078212261199951, "genuineness": 4.469690799713135, "identity": 0.5775383710861206, "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/casual_not_formal/0d5b17b7a6aa84b43f9bcbfedcd9e45c4702cc6679b0b987c7b993ed4d11b36c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "0d83b52fc3f0ec8f9c2a6021fd03bddd1524e8025a5ddbcca85d2ffd2c958286", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1aba799442d9b46d7f3772bbe91c060b5bc093006f270359151580e44cac9969", "generation_seconds_batch": 24.41993298009038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089384078979492, "target_z": 0.12255860351107499, "ce": 5.919358730316162, "genuineness": 4.394378662109375, "identity": 0.7427414059638977, "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/casual_not_formal/0d83b52fc3f0ec8f9c2a6021fd03bddd1524e8025a5ddbcca85d2ffd2c958286.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "0e01b4435e80032fc755927a6595e5c975c8a9540ba9c7f08d07726b4bf661c3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41ac601bb10da1b9510807f7c68dee3c7fe8b92d502a07afa582adb3125308f2", "generation_seconds_batch": 21.978913510916755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1125242710113525, "target_z": -0.6035156094928903, "ce": 6.149696350097656, "genuineness": 4.65797233581543, "identity": 0.6181079745292664, "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/casual_not_formal/0e01b4435e80032fc755927a6595e5c975c8a9540ba9c7f08d07726b4bf661c3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "0e2fb7c46bc54d059505f6100bf4e9191693e0819f5dbf281232b4fe858087bc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "200346d0106e2a531a2f4c4985a76eb810d66945cf52e0a749ec3a9b842a8a64", "generation_seconds_batch": 21.248646283056587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0676684379577637, "target_z": -0.6699218858865901, "ce": 5.808161735534668, "genuineness": 3.9048471450805664, "identity": 0.5445239543914795, "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/casual_not_formal/0e2fb7c46bc54d059505f6100bf4e9191693e0819f5dbf281232b4fe858087bc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "0e786d86b5cb8442fa9ece3eb51106a09d02913a267858b8d6c99d978eeb08f9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a726209e9efea7e60ee3185ace83346536c9d1bd1eb732a05e0229d61f5ccb1e", "generation_seconds_batch": 25.066753593971953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.790968418121338, "target_z": -0.09936521303841749, "ce": 5.998785495758057, "genuineness": 4.369274139404297, "identity": 0.718923032283783, "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/casual_not_formal/0e786d86b5cb8442fa9ece3eb51106a09d02913a267858b8d6c99d978eeb08f9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "0e852f982a20c785b35d4c24afc0e2a4dd740534962b736f852635453159fb65", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cbf2a34376745d5a8d840e680161f8fcd9c2b9618a35408aaa238ffc5421004c", "generation_seconds_batch": 25.066753593971953, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.8828124854541306, "ce": 5.808161735534668, "genuineness": 4.319066047668457, "identity": 0.5042651891708374, "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/casual_not_formal/0e852f982a20c785b35d4c24afc0e2a4dd740534962b736f852635453159fb65.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "0edafae5ce6fe959f8f7a73e680dd62c4d901fce925a3c7e22c626ab056f9607", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a36055051c4b4aaf963056600a98cf7017c95582a5ffb519cb6c9874d59be678", "generation_seconds_batch": 16.321908641955815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4701247215270996, "target_z": -0.38720701472482405, "ce": 6.022613525390625, "genuineness": 4.770940780639648, "identity": 0.7255083918571472, "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/casual_not_formal/0edafae5ce6fe959f8f7a73e680dd62c4d901fce925a3c7e22c626ab056f9607.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "0ee4feb0b32c1cf29b5ad03e0cedbdf7c89f6e7e9b9eaf20962d1701c6256cb4", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2e1a710b498c405eb8a744959e854c84b55f4d79bf4674b4e94a8ff1c73ecb1", "generation_seconds_batch": 17.66224496497307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2844715118408203, "target_z": -0.5849609501201849, "ce": 6.133810997009277, "genuineness": 4.394378662109375, "identity": 0.5334327816963196, "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/casual_not_formal/0ee4feb0b32c1cf29b5ad03e0cedbdf7c89f6e7e9b9eaf20962d1701c6256cb4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "0ef3882364772a07dc6a78259035dc8633b08e5814bea7f62f998b6e16d31894", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d70799879caa6ec522eb80bc2382b3e0697b7f8ac4a043659a8f0c52c106517c", "generation_seconds_batch": 24.746613061986864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.378454685211182, "target_z": 0.3671874314229607, "ce": 6.197352409362793, "genuineness": 4.532451152801514, "identity": 0.5040777921676636, "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/casual_not_formal/0ef3882364772a07dc6a78259035dc8633b08e5814bea7f62f998b6e16d31894.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "0f8f86d27af8553c50defed101124552a682b4c19c62905528150803a9c93af0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19927b58bffd85e956d23adaa2427c13db6de97cb3b20a19cf6a4d928f626d43", "generation_seconds_batch": 21.837526100920513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9430689811706543, "target_z": -0.8476562676016682, "ce": 6.117925643920898, "genuineness": 4.507347106933594, "identity": 0.5809594392776489, "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/casual_not_formal/0f8f86d27af8553c50defed101124552a682b4c19c62905528150803a9c93af0.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": "0fb10140c45dd30b9d307a90b38cb32374b212e6a7f907a1fb6841c2957f7dca", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "2fff9cec32b40f62d9c38a8fe563bf2ac4c6febbcba9ad3b9eb564e15daa665a", "generation_seconds_batch": 16.116155511001125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8277251720428467, "target_z": -0.21093750947274248, "ce": 6.24500846862793, "genuineness": 4.306513786315918, "identity": 0.6722736358642578, "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/casual_not_formal/0fb10140c45dd30b9d307a90b38cb32374b212e6a7f907a1fb6841c2957f7dca.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "1088cc6780e306e35b1f663eb7912661f7914d62aa8e217f23c138f3bf12c40d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53d85c32d10fbad6edea76364aecbad8b32c9b5bd09fca8c5f891f1eacb52360", "generation_seconds_batch": 25.368043602094986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.428294658660889, "target_z": 0.5273437725239261, "ce": 6.173524379730225, "genuineness": 4.300237655639648, "identity": 0.301044762134552, "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/casual_not_formal/1088cc6780e306e35b1f663eb7912661f7914d62aa8e217f23c138f3bf12c40d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "10ac873c9426ac2965749837b22fb8d4e1ead20edfab4220cad71f2444516515", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1fd6a768d8219fdcfa2f46dc84a8f6abcba18eeaf26275c5936976b11c37bdf", "generation_seconds_batch": 17.611180773936212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.8085936995960932, "ce": 6.467403411865234, "genuineness": 4.494794845581055, "identity": 0.2861805856227875, "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/casual_not_formal/10ac873c9426ac2965749837b22fb8d4e1ead20edfab4220cad71f2444516515.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "10cf68137cc118bb285133f3e17d1764db64c834a70cdd452f44abef0370ce8a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "65a9f42b66d26db12f09c0981db1a1bdd027def60079c2aa144e6186baae8eba", "generation_seconds_batch": 23.887470773071982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135664463043213, "target_z": -1.3281249705496978, "ce": 6.324435234069824, "genuineness": 4.444586753845215, "identity": 0.7125774621963501, "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/casual_not_formal/10cf68137cc118bb285133f3e17d1764db64c834a70cdd452f44abef0370ce8a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "1102ae5322753c994f30997756e2ae9d1e331b2ff4751b522ab5b97dff802997", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07476ec3ec04534ba4dd4336a75809d85cae076db69cd0d8fbd883dfe8f4fbca", "generation_seconds_batch": 24.59152119990904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.876497507095337, "target_z": -1.656249921135046, "ce": 6.403861999511719, "genuineness": 4.13078498840332, "identity": 0.6271480917930603, "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/casual_not_formal/1102ae5322753c994f30997756e2ae9d1e331b2ff4751b522ab5b97dff802997.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "11216dc86fbaeb2e5b661677811c952fd760609dd9432237d191303b101ec074", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb073ffbab5497823fe0b5716fbc017abdd295997a7d37e20a05daa8e007c59e", "generation_seconds_batch": 18.310392709099688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.567490816116333, "target_z": -1.7109374848426508, "ce": 6.23706579208374, "genuineness": 4.243753433227539, "identity": 0.5571374893188477, "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/casual_not_formal/11216dc86fbaeb2e5b661677811c952fd760609dd9432237d191303b101ec074.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "11248058b6a9d842e9d99e5924a673555f4beb1608651d85bba15af21336c626", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "27ad1bc93b1fc196df752f4fd5a1c0404cd0afbd91bbb0e0669eb369eb4cab49", "generation_seconds_batch": 25.618976086960174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.5605468887006817, "ce": 6.038498878479004, "genuineness": 4.3127899169921875, "identity": 0.5318529605865479, "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/casual_not_formal/11248058b6a9d842e9d99e5924a673555f4beb1608651d85bba15af21336c626.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "112a05bce2b636e078547274d811c4283533207129b3cbc7307b38af8a61c672", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bbca26bc8214f4a5be73815ceb1f46ec2bda4fedf15ba88d5a12b7b334ed56d", "generation_seconds_batch": 23.068853755947202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1556003093719482, "target_z": -1.0507812761896054, "ce": 5.990842819213867, "genuineness": 4.143336772918701, "identity": 0.6114910840988159, "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/casual_not_formal/112a05bce2b636e078547274d811c4283533207129b3cbc7307b38af8a61c672.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "11648e5f054f5854b7e84fb19d93f028e7a037dd8154201adf33c18e43f9771e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "741f83b076137f7fcbe15171e7644072ea3cc8a37886c1593a53e041b680e818", "generation_seconds_batch": 17.96715675597079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.285183906555176, "target_z": -1.0664061562770328, "ce": 6.268836498260498, "genuineness": 4.394378662109375, "identity": 0.7210950255393982, "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/casual_not_formal/11648e5f054f5854b7e84fb19d93f028e7a037dd8154201adf33c18e43f9771e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "119762d41aef0ca33e68f7ec8bd1a3f1280c97044c02b4fc4ff19b227e6fe22b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8050c9adef864b8e003e84791bd53d52cf43b8eb71d0f00f686c274fca070d03", "generation_seconds_batch": 18.66885324800387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9161908626556396, "target_z": -0.02001950719297073, "ce": 6.340320587158203, "genuineness": 4.306513786315918, "identity": 0.25953108072280884, "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/casual_not_formal/119762d41aef0ca33e68f7ec8bd1a3f1280c97044c02b4fc4ff19b227e6fe22b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "11e95cdc0aebc20a61f81525bb55fbad191297183cd4d36e1c0637fee6123fe2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac7eb8ad6a2ef71e00feb6cc3251f3d072d85c4cf41c7a0aad8f335190d13cc4", "generation_seconds_batch": 22.075880830991082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7274227142333984, "target_z": -0.03637690268067349, "ce": 6.086154937744141, "genuineness": 4.068024158477783, "identity": 0.16922983527183533, "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/casual_not_formal/11e95cdc0aebc20a61f81525bb55fbad191297183cd4d36e1c0637fee6123fe2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "11ea4a6e89ee432a6f374649e9376b1b194b73007659a14e47384538c5adacd9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbff36473805018db11da6f4df7cc3cfb499b84deb311e4cea26f10b7ded8389", "generation_seconds_batch": 16.529907124931924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.472616672515869, "target_z": -0.2543945013783063, "ce": 6.078212261199951, "genuineness": 4.300237655639648, "identity": 0.7021873593330383, "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/casual_not_formal/11ea4a6e89ee432a6f374649e9376b1b194b73007659a14e47384538c5adacd9.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": "1256ee0ac3d2e2503a5454ff3dd116f795937d979705f2f1e21d846625f4b50f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8ea80df796220973b9c89825120964083e2f1ffcf0a010095b392ca965ac5661", "generation_seconds_batch": 18.534652152913623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7691633701324463, "target_z": -0.2641601383233828, "ce": 6.117925643920898, "genuineness": 4.394378662109375, "identity": 0.7828028798103333, "hypothesis": "Ik schreef het nummer op de envelop. Lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/1256ee0ac3d2e2503a5454ff3dd116f795937d979705f2f1e21d846625f4b50f.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "101b145fb36a6df0826a3d7ea6c4bf87b371d264e4109d0c6f908d024d73c360", "generation_seconds_batch": 16.16228816693183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135664463043213, "target_z": -1.1230468801855757, "ce": 5.998785495758057, "genuineness": 4.432034492492676, "identity": 0.6988233923912048, "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/casual_not_formal/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "12c77e26b661c7cc6e7ccdf05fe0339d85c8e7935ab39d3ff7f55aa6710b1382", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "aebf45cae33e15975377c3d41a8464c2df58361dec33f4ad1dc91e47077eda7d", "generation_seconds_batch": 22.768299942021258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3667070865631104, "target_z": -0.5605468821760904, "ce": 6.173524379730225, "genuineness": 4.281409740447998, "identity": 0.5420850515365601, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/12c77e26b661c7cc6e7ccdf05fe0339d85c8e7935ab39d3ff7f55aa6710b1382.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "12c8404b9873b8656bc89552c2a176fb46efed243df32cad0104aab6e32820e2", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e3a7f009afd555f5b9513ba522cffa9093134ee13e9af6a1d2caf8b338d2298", "generation_seconds_batch": 16.542153516900726, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.887532949447632, "target_z": -0.14355467509671357, "ce": 6.292664527893066, "genuineness": 4.1998209953308105, "identity": 0.2732902765274048, "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/casual_not_formal/12c8404b9873b8656bc89552c2a176fb46efed243df32cad0104aab6e32820e2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "12d61c08232727a871466b871671ff9e6fdbc1e9899288eb106e8fb8fd612505", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da0bb2c0e8a73846febaab6cbf9fa33fe1b0068bd18d4842497e14eaf14b7de7", "generation_seconds_batch": 21.827250136062503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7225942611694336, "target_z": -0.11834720006516529, "ce": 6.062326908111572, "genuineness": 4.444586753845215, "identity": 0.4957605302333832, "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/casual_not_formal/12d61c08232727a871466b871671ff9e6fdbc1e9899288eb106e8fb8fd612505.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "137b9f8a1bed7b4071a7304c37319d084d9186a6dc3d6384586165621d2b08ea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a81ae0568505140363b9d1c1dccbe6d7ddaf95075300636607cba6903d02f9ff", "generation_seconds_batch": 24.447385976091027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -0.8398436996120993, "ce": 6.0305562019348145, "genuineness": 4.432034492492676, "identity": 0.4521660804748535, "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/casual_not_formal/137b9f8a1bed7b4071a7304c37319d084d9186a6dc3d6384586165621d2b08ea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "1399839a5d5277fa765b7927eb2a16fa523332c6f130876cb6216bdca1f66169", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "04106d692fa480b98dbfaaa1656aa15b0a404d6a426b7aa4dc73a004a38430f7", "generation_seconds_batch": 25.36074345104862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9430689811706543, "target_z": -0.5410156015004266, "ce": 6.109982967376709, "genuineness": 4.231201171875, "identity": 0.717089831829071, "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/casual_not_formal/1399839a5d5277fa765b7927eb2a16fa523332c6f130876cb6216bdca1f66169.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": "14c6db701c812dc0817a1ac50d9cb0e01b2243b2b9ce35efc060ad1704b6f371", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 129, "seconds_generated": 10.32, "mp3_sha256": "4707b0652016d2a3f32be4bb61e675d129659763d3823a3db1a60639ba039e78", "generation_seconds_batch": 17.292070416966453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3044073581695557, "target_z": -0.5302734512663113, "ce": 6.157639026641846, "genuineness": 4.444586753845215, "identity": 0.663036584854126, "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": 0.120000000000001, "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/casual_not_formal/14c6db701c812dc0817a1ac50d9cb0e01b2243b2b9ce35efc060ad1704b6f371.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "14f436e34fe7b9d88a30fae6476a4b44ba97aa9666a9f682b3bfc7374126be24", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ec5473ad9d823bc1c750b000731c622430624820179c3e491621df47e92d500", "generation_seconds_batch": 22.530166022013873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269519567489624, "target_z": -0.340820269110509, "ce": 6.133810997009277, "genuineness": 4.268857479095459, "identity": 0.5812070369720459, "hypothesis": "Oh, ik heb het kleine doosje gevonden. 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/casual_not_formal/14f436e34fe7b9d88a30fae6476a4b44ba97aa9666a9f682b3bfc7374126be24.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "15026daf378b373a7651d96526b0bb5137f209ef61981dd552fe930d0b634dbd", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f757ff8432caecbad1a0a9bf0701e84cf06c94a9dc79d4a34a1034cd80fa819", "generation_seconds_batch": 24.4971581940772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.75016188621521, "target_z": -0.035034252403336186, "ce": 6.1020402908325195, "genuineness": 4.20609712600708, "identity": 0.4055030345916748, "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/casual_not_formal/15026daf378b373a7651d96526b0bb5137f209ef61981dd552fe930d0b634dbd.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "1532fae292bbe7c77df3b51c80dff61da6dc9c9e26b83e7ba6013e58746d2dfa", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63139fc9f3ff9a112229f6631d101883e57a48898222a01b5586d9b351788c2b", "generation_seconds_batch": 17.589124253951013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6970741748809814, "target_z": -1.5781250340988324, "ce": 6.205295085906982, "genuineness": 4.262581825256348, "identity": 0.4985227584838867, "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/casual_not_formal/1532fae292bbe7c77df3b51c80dff61da6dc9c9e26b83e7ba6013e58746d2dfa.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": "15347ca5958268a76484d8b7e45b86eb1408e0a8a51f1e690d37914e2f915db4", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "994b402bcf226bb11f9160e3c09273a2563dded2b51d5d987b1b22dc8fbf14e4", "generation_seconds_batch": 18.68792234605644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.6601562245723415, "ce": 6.252951145172119, "genuineness": 4.469690799713135, "identity": 0.7121002674102783, "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/casual_not_formal/15347ca5958268a76484d8b7e45b86eb1408e0a8a51f1e690d37914e2f915db4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "1537982583f5aa6045dfdcfb98468b979d1b41f3ad7e2730b2c7de9f37289509", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1a35894cce9b2906afbba26b6cb3442a7f6f45a44b688b17eceb3656c7078de", "generation_seconds_batch": 23.819262548931874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.87890623682276, "ce": 6.038498878479004, "genuineness": 4.507347106933594, "identity": 0.7271392941474915, "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/casual_not_formal/1537982583f5aa6045dfdcfb98468b979d1b41f3ad7e2730b2c7de9f37289509.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "361b3d99d3c18c667d5c78054555c186197b24e6b0d9971085a62a080de08652", "generation_seconds_batch": 15.995051946956664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.305119514465332, "target_z": -1.32812511630772, "ce": 6.205295085906982, "genuineness": 4.494794845581055, "identity": 0.3888202905654907, "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/casual_not_formal/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "156f78258144473eb32a41a622cd2dc8dad57ddfdd7d588c89387c4c3beb03f5", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f71894d9b6cc161fcb77145396b1cb5209e27168fbfb1f8463dd1419a3157266", "generation_seconds_batch": 16.181671860977076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2844715118408203, "target_z": -0.46386715301539105, "ce": 6.109982967376709, "genuineness": 4.482243061065674, "identity": 0.09473554790019989, "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/casual_not_formal/156f78258144473eb32a41a622cd2dc8dad57ddfdd7d588c89387c4c3beb03f5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "15ae1d5451f4c739402b372992413055a257a9b1816e0728041fa64d02f6412b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "524b178908f6ba7f592ca22404aa943b37a89bbd10ecd509945f8943999f95f2", "generation_seconds_batch": 25.48835280793719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135664463043213, "target_z": -1.414062439146502, "ce": 6.38003396987915, "genuineness": 4.231201171875, "identity": 0.5035635232925415, "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/casual_not_formal/15ae1d5451f4c739402b372992413055a257a9b1816e0728041fa64d02f6412b.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": "15af4763d39c8604f21542180e4d1f5b0719e9314b61da0b7eac9a07f449977f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "b1d6551322c9175453daf359089ee3b4b7343719fb8417dd6408c53d5272010a", "generation_seconds_batch": 16.726159596000798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.457664728164673, "target_z": -0.42919923100361757, "ce": 6.165581703186035, "genuineness": 4.300237655639648, "identity": 0.7790732383728027, "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/casual_not_formal/15af4763d39c8604f21542180e4d1f5b0719e9314b61da0b7eac9a07f449977f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "15b631f211b51d657ed642c3b837659f327450e985cb5ec92706837973e01dc8", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c326bde37d2a000d0dc1b358c1ac9b56537d50ec023c738706e3775daa217e3", "generation_seconds_batch": 18.07277487590909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5792384147644043, "target_z": -1.1386719794318747, "ce": 6.038498878479004, "genuineness": 4.281409740447998, "identity": 0.5542947053909302, "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/casual_not_formal/15b631f211b51d657ed642c3b837659f327450e985cb5ec92706837973e01dc8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "15bffa4986822d7dda9654ba937096e0f821de2bed5693b4fdc2b144b5ab2a91", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ae2b05a37ffb47cb7ec39c3aab0b897d5bcd4fe2575e4c0829c9332c2eb1bc25", "generation_seconds_batch": 16.560599604970776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.4892577563018361, "ce": 5.895530700683594, "genuineness": 4.168440818786621, "identity": 0.7570536136627197, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/15bffa4986822d7dda9654ba937096e0f821de2bed5693b4fdc2b144b5ab2a91.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": "1611d9a1188bfc5ce7db64fdcf9198f0246b44bbf60dfeb4c01d5d03476f2eae", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "289f7a2119a0aae0a27feed24a3f7596a606968800340126fad5cc8244fe0e00", "generation_seconds_batch": 15.98680782399606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5486223697662354, "target_z": -0.34472654658070206, "ce": 6.125868320465088, "genuineness": 4.68307638168335, "identity": 0.739992618560791, "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/casual_not_formal/1611d9a1188bfc5ce7db64fdcf9198f0246b44bbf60dfeb4c01d5d03476f2eae.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "162490406e1574711fdf7a7ceac03a25b46874d260bc925f06c8320f22d2fa31", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "106445d0aa171c8025f0d7ae0475c47cb76d82b19a0d145d9c7693b67744a83d", "generation_seconds_batch": 27.071063847048208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9124526977539062, "target_z": -0.12988287476294436, "ce": 6.316492557525635, "genuineness": 4.582659721374512, "identity": 0.6538964509963989, "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/casual_not_formal/162490406e1574711fdf7a7ceac03a25b46874d260bc925f06c8320f22d2fa31.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "162a1b13d0f06531248f532d8c8eb4f79202ec1f1461c4a581cfc873ebf9f4ad", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "725f386ebf59189790a50f7e8e60d434d8530b890e2fc37dc83fed662cbd3279", "generation_seconds_batch": 21.67579561995808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.888245105743408, "target_z": -0.7929687385395472, "ce": 5.982900142669678, "genuineness": 4.281409740447998, "identity": 0.6342608332633972, "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/casual_not_formal/162a1b13d0f06531248f532d8c8eb4f79202ec1f1461c4a581cfc873ebf9f4ad.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "16db06074618924634da64f98667682ec8cc043c44b6c485fb0643e0afe71506", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "54aa9f64eb3f5c8c30dbda34b1b226fe62dd4cda1f6d6e55b6fe2efb19da0297", "generation_seconds_batch": 24.91230469592847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.929896831512451, "target_z": 0.1252441601296428, "ce": 5.816103935241699, "genuineness": 4.337893962860107, "identity": 0.48016005754470825, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/16db06074618924634da64f98667682ec8cc043c44b6c485fb0643e0afe71506.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "16e7798a5632eda1cda37d67c268641a78ee54446d065607d44a870cb1c4e8e8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0e38226f25031f31e28aa0f409dd5eee87783e9b0425b627a1ecc9a1e38baec5", "generation_seconds_batch": 24.59152119990904, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.51444673538208, "target_z": -1.17187505163327, "ce": 6.427689552307129, "genuineness": 4.406930446624756, "identity": 0.7290956974029541, "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/casual_not_formal/16e7798a5632eda1cda37d67c268641a78ee54446d065607d44a870cb1c4e8e8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "173e220dde494a289ee36db0550d277e4d4d145085b4730fb412b1a94296b3ff", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "518af40cddc86c00a5ee4a2055a7e5fecc728ba685b96c6a33f215f877d6b77e", "generation_seconds_batch": 18.77852300100494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.364927291870117, "target_z": -0.9531250281617114, "ce": 6.221180438995361, "genuineness": 4.137060642242432, "identity": 0.47874537110328674, "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/casual_not_formal/173e220dde494a289ee36db0550d277e4d4d145085b4730fb412b1a94296b3ff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "183a81df6c40528aa63d5015004730b0735e1c54265135260bb5364e38643f0b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4b78973caaf27faf003915fa3d029d512baa1e3a8ea6e0857d75efc47b304ef", "generation_seconds_batch": 23.908204993000254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.336091041564941, "target_z": 0.27539063819284837, "ce": 6.133810997009277, "genuineness": 4.519899368286133, "identity": 0.32655414938926697, "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/casual_not_formal/183a81df6c40528aa63d5015004730b0735e1c54265135260bb5364e38643f0b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "1887431040aff1329bb5e934ec4c5fb24ab2ac8f09cee3fe084f65b512dfed66", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3dfafbb23da69434625fac36e33e1d2f82a7a6c6e4d978bffd21020dbf0c10ac", "generation_seconds_batch": 17.287717034923844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6988539695739746, "target_z": -1.0527343936549518, "ce": 6.086154937744141, "genuineness": 4.2374773025512695, "identity": 0.623943567276001, "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/casual_not_formal/1887431040aff1329bb5e934ec4c5fb24ab2ac8f09cee3fe084f65b512dfed66.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": "191778bc5bb57c73f0bacb8352b0bda17c2a5f65fe978195d3a83a4fba5ac718", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "ed2605cc5fe4ee94a46857e73a74546c9cb2a9d3df3fb4baf1d273fce639f431", "generation_seconds_batch": 17.623624671949074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.573542356491089, "target_z": -0.23437495630996996, "ce": 6.23706579208374, "genuineness": 4.325342178344727, "identity": 0.638579249382019, "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/casual_not_formal/191778bc5bb57c73f0bacb8352b0bda17c2a5f65fe978195d3a83a4fba5ac718.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "192b8714b8afeab2ec9c24f9b8c2a097fd8f0121dc326100117b267cefca9bf9", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a34d3c978be34ee44880ca042e3364156509e521c9280d1b960a446809cb095", "generation_seconds_batch": 18.551569062052295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5493345260620117, "target_z": -1.1386719200590572, "ce": 6.252951145172119, "genuineness": 4.3881025314331055, "identity": 0.7753059267997742, "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/casual_not_formal/192b8714b8afeab2ec9c24f9b8c2a097fd8f0121dc326100117b267cefca9bf9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "198436cb562f369cf38c23c29f35903419e8d65100d0cce44fc0d03a23d09842", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "269224ba5f3fb25459266e01dcfce1eb586a3b073da4e29024daba846460b09e", "generation_seconds_batch": 23.580471798893996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.124271869659424, "target_z": 0.2978515463930156, "ce": 6.149696350097656, "genuineness": 4.6956281661987305, "identity": 0.7380889654159546, "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/casual_not_formal/198436cb562f369cf38c23c29f35903419e8d65100d0cce44fc0d03a23d09842.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "198a79a504fbe41a6ee9cb7df1a3b0d3aaf9d02a511ab5c6d7713d600c6842c8", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94c3767071a3bea4b3bcf44faacbc67555c5f75eec47328ffda53a686968ffac", "generation_seconds_batch": 15.955347063019872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.856561541557312, "target_z": -1.7578124925661924, "ce": 6.427689552307129, "genuineness": 4.193545341491699, "identity": 0.48168039321899414, "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/casual_not_formal/198a79a504fbe41a6ee9cb7df1a3b0d3aaf9d02a511ab5c6d7713d600c6842c8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "199e4e742ebee52d58c3ff5c32726d7e80a6c85d960030d84e6ef2dc13fcd48d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "179ecf0395f4e81678ed8b404d1c00a71e8e43a673a4129451270307d2f8ca87", "generation_seconds_batch": 23.17944325506687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.247091770172119, "target_z": -0.32519524806045763, "ce": 6.181467056274414, "genuineness": 4.118232727050781, "identity": 0.44188183546066284, "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/casual_not_formal/199e4e742ebee52d58c3ff5c32726d7e80a6c85d960030d84e6ef2dc13fcd48d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "19d57e1c18bda0433e86509444daa919a2bbec3ab7200c8a8fd6473c1cfdc48a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20e04af24d7b4e42411925cb4574f8d1f56d382da17afdb15e64a7347a3be1f6", "generation_seconds_batch": 22.465436334023252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.98614501953125, "target_z": -1.5976562497246736, "ce": 6.38003396987915, "genuineness": 4.394378662109375, "identity": 0.47020721435546875, "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/casual_not_formal/19d57e1c18bda0433e86509444daa919a2bbec3ab7200c8a8fd6473c1cfdc48a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "19dc90301316a5062ab82ff42d862faef883a0bf311af4d250c054bc02c732a8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b50d5bb1369cbc4f661a40275cc1130e5b8a6be8b7eca4e895dc8f75bea072b2", "generation_seconds_batch": 27.071063847048208, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9093377590179443, "target_z": -0.17016606091656356, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.6823780536651611, "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/casual_not_formal/19dc90301316a5062ab82ff42d862faef883a0bf311af4d250c054bc02c732a8.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": "1a9fc5f7123436911b7f8c3b1e0fe6242070ad4ecf02a3654b023484a03540aa", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7099fd20d126d48fd3b45ed7b79de0056ba87c7a920c2fe86f6cee5bea01f743", "generation_seconds_batch": 16.13566301204264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.338582992553711, "target_z": 0.3398437289727606, "ce": 6.340320587158203, "genuineness": 4.570107460021973, "identity": 0.537659764289856, "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": 0.03999999999999915, "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/casual_not_formal/1a9fc5f7123436911b7f8c3b1e0fe6242070ad4ecf02a3654b023484a03540aa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "1b1222c369b3f36d342429bc6ef70415806b450b412aef6d0566dd41b4daf9b0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94fe854d2512161a4f55ad355f325999920c06d86896efe4a6fed10b195bb2af", "generation_seconds_batch": 23.81996055901982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823453426361084, "target_z": -0.8671875470375632, "ce": 5.824047088623047, "genuineness": 4.068024158477783, "identity": 0.7705352306365967, "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/casual_not_formal/1b1222c369b3f36d342429bc6ef70415806b450b412aef6d0566dd41b4daf9b0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "1b2195a4081a3ba037896a0bc5c8708c3ecaa89b4e8da459663a26e124ba691b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f11385de7aaa566794ecd10513f8cd7f627ca6eec96cee90dee321ada1722fe0", "generation_seconds_batch": 24.141512085916474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.346771240234375, "target_z": -0.5097656319535873, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.6735508441925049, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/1b2195a4081a3ba037896a0bc5c8708c3ecaa89b4e8da459663a26e124ba691b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "1b30b38e8b3b7e7ae0881826d2195c0f3674e841aaa8b3bf108b9f03d3647672", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff68a903e1d3c7d1b87edc6de245bb6ad5f242d21bdeaa5e6f922b97fbe87fbc", "generation_seconds_batch": 24.74133886594791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1997437477111816, "target_z": -0.4208984983815108, "ce": 6.284721851348877, "genuineness": 4.419482707977295, "identity": 0.1745387315750122, "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/casual_not_formal/1b30b38e8b3b7e7ae0881826d2195c0f3674e841aaa8b3bf108b9f03d3647672.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": "1b5a7aad85fba80fc5d40d8f9a12a1906e400caf710a8020c48962aa4cde398b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f41ae1a9201aa216a0586d87236010d2c70d2635ac7eb9052dac717116d830ab", "generation_seconds_batch": 16.02424073696602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8950088024139404, "target_z": 0.041992132340712585, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.770377516746521, "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/casual_not_formal/1b5a7aad85fba80fc5d40d8f9a12a1906e400caf710a8020c48962aa4cde398b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "1c4d8417c2c85462b833ad8f632e0f7af391dafb55f4c94bf6b0823ba11a1430", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "68352b68aa992dffa59f1e6ea572e384cfd23dad83169a998cb38df85d06e8a1", "generation_seconds_batch": 17.62405568896793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1075401306152344, "target_z": -0.5839844011095707, "ce": 6.332377910614014, "genuineness": 4.375550270080566, "identity": 0.35252779722213745, "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/casual_not_formal/1c4d8417c2c85462b833ad8f632e0f7af391dafb55f4c94bf6b0823ba11a1430.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "1c6dcc9b2fb4726e1244af59d777801a5d8c95bf56f0f604030c17db751281d7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b3398383e3031462cdc22e9fc696fa78f57f60dfd2fcae5d8575714655492508", "generation_seconds_batch": 23.590502972947434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1573801040649414, "target_z": -0.6542968977528292, "ce": 6.0305562019348145, "genuineness": 4.394378662109375, "identity": 0.5614103078842163, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce que le", "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/casual_not_formal/1c6dcc9b2fb4726e1244af59d777801a5d8c95bf56f0f604030c17db751281d7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "1c96e007a7c08c086e1470a583262e4b30d0b38dab58d18ed64518aaa72bbb89", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "36cb01fe54a70f1a07f0da412317c2959995565f886745f50630f972c7eeb590", "generation_seconds_batch": 23.395032897009514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.759506940841675, "target_z": -0.05895998357367719, "ce": 5.77639102935791, "genuineness": 4.281409740447998, "identity": 0.6853468418121338, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klassa.", "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/casual_not_formal/1c96e007a7c08c086e1470a583262e4b30d0b38dab58d18ed64518aaa72bbb89.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "1d07ffb07b99af1b1345c60850127994a8837938297f47481a8ef866ebf75cec", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "80ce62c7072cd0fb435ba862e142eba6c9eec14122fd25733b7e3d1ce90df6b5", "generation_seconds_batch": 17.162246962077916, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2445995807647705, "target_z": -0.6142578622275973, "ce": 6.109982967376709, "genuineness": 4.507347106933594, "identity": 0.7453517913818359, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/1d07ffb07b99af1b1345c60850127994a8837938297f47481a8ef866ebf75cec.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "1d29b6cff56d7ced57b1ed63d26668868dd78b50c0b655f08e5103ff081756f2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0739f07768cae052102f2cc46a00550e055a00ecdb22ad6bba9e4f3e31004bce", "generation_seconds_batch": 23.6127446340397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1623640060424805, "target_z": -0.6406249872853712, "ce": 6.284721851348877, "genuineness": 4.306513786315918, "identity": 0.7518491148948669, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/1d29b6cff56d7ced57b1ed63d26668868dd78b50c0b655f08e5103ff081756f2.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "07666e49d5bdaf3b6de9b79fd8a671d2d88abf1eb925a4a70b9ec02340a0b664", "generation_seconds_batch": 16.2006766180275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6572024822235107, "target_z": -1.8828124819309155, "ce": 6.252951145172119, "genuineness": 3.9927117824554443, "identity": 0.7631918787956238, "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/casual_not_formal/1d559e6da25f8bbacc3b53d5880369445633ce7ab0125ca7b07093108b8c8de3.mp3" }, { "id": "heldout-de-1", "language": "de", "sentences": [ "Die Schlüssel liegen auf dem Küchentisch.", "Ich suchte in der Schublade... dann neben dem Fenster.", "Nimm sie mit, wenn du gehst." ], "text": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade... dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-de-1", "key": "1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "3b88f69bd0211b60e96b6971f51ac0402519050c199f10c7fbf66d9c1bcbf20f", "generation_seconds_batch": 16.548972547985613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5742545127868652, "target_z": -1.0781250004321978, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.6959083080291748, "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/casual_not_formal/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "1da8647ff1f1835b84cc38e81bce056124eb9437eb2a0592e24fb710bcb5c4ce", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b96f7809c28259dd5985a383504600a3c4d182a83bd5d24197f51ba261d933f", "generation_seconds_batch": 26.79759092105087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.4970703308159965, "ce": 6.006728172302246, "genuineness": 4.306513786315918, "identity": 0.5803686380386353, "hypothesis": "La puerta sigue abierta. Y conoce de que no ciervú. He dejado tu abrigo sobre la silla. Dime cuándo estás.", "wer": 0.4666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/1da8647ff1f1835b84cc38e81bce056124eb9437eb2a0592e24fb710bcb5c4ce.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "1da9ef013fd9bb762ccaa543de570430792c8ceea61b0a70219f8176e44487be", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93d156fe04fbf966175390f7079903323dc7bc3b4cc87143148357d8bc5a7e3b", "generation_seconds_batch": 17.855723098036833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.344991445541382, "target_z": -1.2539062354525314, "ce": 6.229123115539551, "genuineness": 4.337893962860107, "identity": 0.4562429189682007, "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/casual_not_formal/1da9ef013fd9bb762ccaa543de570430792c8ceea61b0a70219f8176e44487be.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "1e237f3585707218f54c8ee11aa99cf21eed57112d9f92e4c81c5070080c8f5b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "49ea4308e1b137fec6b445c43347e827797047f00cff75bb689b48f88d1a9ef2", "generation_seconds_batch": 27.056125040980987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.933100938796997, "target_z": -0.6953125434247074, "ce": 6.149696350097656, "genuineness": 4.068024158477783, "identity": 0.8157332539558411, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin!", "wer": 0.125, "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/casual_not_formal/1e237f3585707218f54c8ee11aa99cf21eed57112d9f92e4c81c5070080c8f5b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "1e266a035d32499d390841b5ddd019a2642577260ae9a85afb60828be9c1447e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "735b82eda5c310d30a42d20df3b4695ab4bec7bdc82b6943ef247ff22a056c25", "generation_seconds_batch": 18.437574377050623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.614126205444336, "target_z": -0.9999999805574891, "ce": 6.292664527893066, "genuineness": 4.444586753845215, "identity": 0.6222439408302307, "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/casual_not_formal/1e266a035d32499d390841b5ddd019a2642577260ae9a85afb60828be9c1447e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "1e5580122d5f65c3d2a26a16e1fb13332d0d33d2578938cb2f1bf07c15362ba6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a321963d72040b76cc1ad857e3c08e05cd175df390fd18d1dae4f80ea6714b1", "generation_seconds_batch": 25.856807679054327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.749227523803711, "target_z": -0.011230442519174955, "ce": 6.022613525390625, "genuineness": 4.2751336097717285, "identity": 0.6155168414115906, "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/casual_not_formal/1e5580122d5f65c3d2a26a16e1fb13332d0d33d2578938cb2f1bf07c15362ba6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "1e55c5f19e3ac692116226e9d3c99cd0f270c8a3630b770c5e1735917fe4b704", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ebbca10211eb90f1da194ae779a6024ab81bfa05651ca8ae0a5bde01aafa8a2", "generation_seconds_batch": 22.435147687094286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.648301839828491, "target_z": -0.24804689352889545, "ce": 6.340320587158203, "genuineness": 4.444586753845215, "identity": 0.20063543319702148, "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/casual_not_formal/1e55c5f19e3ac692116226e9d3c99cd0f270c8a3630b770c5e1735917fe4b704.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "1e668a157f141f59ecb108c959fd641ca251a22da2114a4e45cc14cbcaf32989", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "90462bb2f086b9e7bc8e1be8d8c56faee6c6789cc45606e749df392bc12d200a", "generation_seconds_batch": 25.88203698990401, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6988539695739746, "target_z": -0.9394530614524685, "ce": 6.24500846862793, "genuineness": 4.582659721374512, "identity": 0.7246392369270325, "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/casual_not_formal/1e668a157f141f59ecb108c959fd641ca251a22da2114a4e45cc14cbcaf32989.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "c72138ad1c9b5e453c3e1aabd639367facbd03bf2ad1ad0ae26f1eeb3f474bf7", "generation_seconds_batch": 16.2006766180275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.285183906555176, "target_z": -1.2851561193320964, "ce": 6.268836498260498, "genuineness": 4.419482707977295, "identity": 0.6645997762680054, "hypothesis": "Te escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame, cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "1ed8880ed25a093cd9117b7d09c1e8daaa479e4b0135d491f42c4a9719f5555f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49c2ab43fefd2320030329271d36583ff80eeb82842440c7bb9ee0f4c07b2e50", "generation_seconds_batch": 17.287717034923844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.504478931427002, "target_z": -1.097656119495372, "ce": 6.109982967376709, "genuineness": 4.400654315948486, "identity": 0.7112700939178467, "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/casual_not_formal/1ed8880ed25a093cd9117b7d09c1e8daaa479e4b0135d491f42c4a9719f5555f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "1f0d33674f53e30ec6b80e6cd1fb4a147eb6f6ae706d18b8fff169895113c639", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f775309bf190ccc11d3bc4f838968f964f81908a60c61e3345c9bd941aed6b0", "generation_seconds_batch": 19.106700982898474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5774586200714111, "target_z": -1.5742188006039997, "ce": 6.173524379730225, "genuineness": 3.9425034523010254, "identity": 0.5692741870880127, "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/casual_not_formal/1f0d33674f53e30ec6b80e6cd1fb4a147eb6f6ae706d18b8fff169895113c639.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "2027e5a79b26e7ec10351c296daf73ab177196c92254f87b3f01d6e155c38566", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93b856a5208387064a026f96676667f2c6952513d8f59da133d628fe6d2a99f1", "generation_seconds_batch": 21.607974275015295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.464607000350952, "target_z": -1.058593668148077, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.5847071409225464, "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/casual_not_formal/2027e5a79b26e7ec10351c296daf73ab177196c92254f87b3f01d6e155c38566.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "203a9a491aebd3d10c5af81c31acf95800c76964053b5a5f4dd5130086bc85cb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bc2acc16e9e51262d837cfce1e6b80c0799e160ee0b10abae50f30727832d8c", "generation_seconds_batch": 26.06323558499571, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.7734375039265987, "ce": 6.308549880981445, "genuineness": 4.419482707977295, "identity": 0.45285505056381226, "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/casual_not_formal/203a9a491aebd3d10c5af81c31acf95800c76964053b5a5f4dd5130086bc85cb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "204b59d6f8df0ac798012b5f2b1cee9bccf3713cad856521557c059460082820", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51df859fe0aebbfc63d5b440018d3903ca86b215a7861729575c82a0fb5a8dae", "generation_seconds_batch": 22.654940251028165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3492631912231445, "target_z": -0.538086005227848, "ce": 6.284721851348877, "genuineness": 4.507347106933594, "identity": 0.6602919101715088, "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/casual_not_formal/204b59d6f8df0ac798012b5f2b1cee9bccf3713cad856521557c059460082820.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "2078637fba0baf777770b94e4843e8e0716c17ed9a6e6e12ed0870349d7d482a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b219af40142176dad51bbd176ca641e53d6726e9eef3eea43e2a19a64822085", "generation_seconds_batch": 21.420616014976986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9598007202148438, "target_z": -0.0546874925902015, "ce": 6.149696350097656, "genuineness": 4.607763767242432, "identity": 0.7581509351730347, "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/casual_not_formal/2078637fba0baf777770b94e4843e8e0716c17ed9a6e6e12ed0870349d7d482a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "207f2bf474c13f0a6101163359837fa1ecc0d7b248e2e44d32ace00e18ba6333", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3ffde18b2c5b4c07bfe2072536ac685f46495534d8ddd86ff713d1a0464a3757", "generation_seconds_batch": 23.134691886953078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2769954204559326, "target_z": -0.35839849492352555, "ce": 5.816103935241699, "genuineness": 4.457139015197754, "identity": 0.6113038063049316, "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/casual_not_formal/207f2bf474c13f0a6101163359837fa1ecc0d7b248e2e44d32ace00e18ba6333.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "21b8dd8e84b1683153a41aab0be90678f99c4004b40bf9224345b6dee8c0b5b9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dad75eb0b36b0241184d1da9891f39ef121f77ccd1bf04a84956878a01979d0f", "generation_seconds_batch": 23.58782120095566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6339728832244873, "target_z": -0.1911621199918687, "ce": 6.09409761428833, "genuineness": 4.319066047668457, "identity": 0.7619619369506836, "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/casual_not_formal/21b8dd8e84b1683153a41aab0be90678f99c4004b40bf9224345b6dee8c0b5b9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "21bb4ca48241434ffa3999c69095bc0d5dd3c584d742c90e3b7a52f73f7ba246", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fd3c114d7e3c40e97642b2e8b219bf70c0e675d96d479be7bd3c982fee46141", "generation_seconds_batch": 24.657570970011875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4589107036590576, "target_z": -0.11181644577351893, "ce": 6.046441555023193, "genuineness": 3.9550557136535645, "identity": 0.3305011987686157, "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/casual_not_formal/21bb4ca48241434ffa3999c69095bc0d5dd3c584d742c90e3b7a52f73f7ba246.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "21bd1562fb315201ea7b26f74842fe4c84a16c03916bdf3f4889010acea34a7d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88ce8944743a74104aa94f4c7ff7a9b387f7b45a917df0e919e9c2f9d0509ac9", "generation_seconds_batch": 16.560599604970776, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.987924814224243, "target_z": -0.6347655649468693, "ce": 6.173524379730225, "genuineness": 4.218649387359619, "identity": 0.5636382699012756, "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/casual_not_formal/21bd1562fb315201ea7b26f74842fe4c84a16c03916bdf3f4889010acea34a7d.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": "21d00db134f3a35215b8d41bf96d15b44b5ad82b60d7009ee9ff0b1329ff46c0", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1270af9ca17d7b4481a951e80005c0f5f7372c175ba8e930bd86ff8ce1767039", "generation_seconds_batch": 18.20180156605784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3617231845855713, "target_z": -0.5058593700769427, "ce": 6.221180438995361, "genuineness": 4.582659721374512, "identity": 0.5796892046928406, "hypothesis": "De sleutels liggen op de keukentafel. Ik kijk in de la 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/casual_not_formal/21d00db134f3a35215b8d41bf96d15b44b5ad82b60d7009ee9ff0b1329ff46c0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "21e946e3a1218ed5a7e36d67bc3fcb86e50171b740567873309a60fb36c992b6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ceb0d3516f9dcee2930ac9e151dab651cce8960dfeac0827fc3c63e0cb1321c1", "generation_seconds_batch": 24.388761150068603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9630048274993896, "target_z": -0.6250000048934437, "ce": 6.23706579208374, "genuineness": 4.074300289154053, "identity": 0.2370631843805313, "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/casual_not_formal/21e946e3a1218ed5a7e36d67bc3fcb86e50171b740567873309a60fb36c992b6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "21ead25a6d30acad545ddcf0f430e6702713a236451a21fb8ae595b57d30d857", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7690d59620c3f071c615813f2cd2019ad5d8699aaaa1f26eaca9924f3503ca30", "generation_seconds_batch": 24.74976805096958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7985334396362305, "target_z": -0.5136719352612267, "ce": 6.205295085906982, "genuineness": 4.381826400756836, "identity": 0.5777127742767334, "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/casual_not_formal/21ead25a6d30acad545ddcf0f430e6702713a236451a21fb8ae595b57d30d857.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "2259ca00acebc6de364249b7e926c198cacda4c103a761bcca17b58a48becef7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "79e872dcfa24ff59f87f99e62e708657736ff556082f3154df93ed28fe660fe8", "generation_seconds_batch": 24.41254854190629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2445995807647705, "target_z": -0.2880859747586812, "ce": 6.070269584655762, "genuineness": 4.369274139404297, "identity": 0.45567408204078674, "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/casual_not_formal/2259ca00acebc6de364249b7e926c198cacda4c103a761bcca17b58a48becef7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "225ebff9fbb62a9f32078ad7d585b2f4f0dd79ce1e94dea56ddfeef2d82e76c7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf5f019e0b094aa00b54417d29fa10814a49bbbd2a85699f95225108c9018611", "generation_seconds_batch": 18.291450777091086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7437098026275635, "target_z": -0.8476562741262597, "ce": 6.252951145172119, "genuineness": 4.557555198669434, "identity": 0.5495255589485168, "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/casual_not_formal/225ebff9fbb62a9f32078ad7d585b2f4f0dd79ce1e94dea56ddfeef2d82e76c7.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": "22f3595d1118b8b1e96caab4bfd4f29b0787e5f54f4af0132a079e343f9d9e8f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "2b83b7af1fd50901fbfcb76ce8b35bd13afe61e6648e510089ca04bc590dfa06", "generation_seconds_batch": 17.15244798909407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.218967437744141, "target_z": 0.2626953139906823, "ce": 6.316492557525635, "genuineness": 4.055472373962402, "identity": 0.426056444644928, "hypothesis": "The keys are on the kitchen table. I checked the drawer, then looked aside the window. Take them with you when you leave.", "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/casual_not_formal/22f3595d1118b8b1e96caab4bfd4f29b0787e5f54f4af0132a079e343f9d9e8f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "23152e13bf687beae1452fa948cf5353a79e96dff98b995d5ecfcf048702a534", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5cca8681a90e85ab28a5c572a0d04d9866397dd92d88fd192f7933551f55013b", "generation_seconds_batch": 23.30655007099267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.538654327392578, "target_z": -0.3974609718797615, "ce": 6.316492557525635, "genuineness": 4.645420074462891, "identity": 0.8159195184707642, "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/casual_not_formal/23152e13bf687beae1452fa948cf5353a79e96dff98b995d5ecfcf048702a534.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "23cda104f1435dd7dc220d75dbc6c8fc90158dfcaf1a26ba86ee8f508b502f0a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8df09d2318069514883d25efa6650ae1b2273c55e7d8c2eda5b5ff100b785519", "generation_seconds_batch": 23.49493841896765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.268807411193848, "target_z": 0.3906250304779735, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.8006100654602051, "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/casual_not_formal/23cda104f1435dd7dc220d75dbc6c8fc90158dfcaf1a26ba86ee8f508b502f0a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "23f81c775553995abcc79b6f857fe3e2e017ec104e3d3f11efb96dae435b02bb", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b1a8c6e2ff6ac5b2bbc05aa1bed37cdc33da87d03d62f98e985fcd0977930bb", "generation_seconds_batch": 17.986038436065428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.5507812339110245, "ce": 6.046441555023193, "genuineness": 4.293961524963379, "identity": 0.53645920753479, "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/casual_not_formal/23f81c775553995abcc79b6f857fe3e2e017ec104e3d3f11efb96dae435b02bb.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": "24517f2ed98799b261c48138125553cfcf26332d8d12295506ce5c15a14f78e4", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a7bcba32b9c87bcdaa03e337d3da780a9603c6d489ed44be340d0d524e4e486", "generation_seconds_batch": 24.307468662038445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.69315767288208, "target_z": -0.18066408857835595, "ce": 5.998785495758057, "genuineness": 4.325342178344727, "identity": 0.6998313665390015, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten. 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/casual_not_formal/24517f2ed98799b261c48138125553cfcf26332d8d12295506ce5c15a14f78e4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "2478ece120c5bf5c87a4eabbc2d38ef04aec4242d60f451d3f775b33d6091fa5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3196b368a8fc7e61eb7d3f6421b084c598cdc3894531bd0d1c222330fefefb6", "generation_seconds_batch": 27.12334181694314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042036056518555, "target_z": 0.10778799470922598, "ce": 6.1020402908325195, "genuineness": 4.1998209953308105, "identity": 0.6927175521850586, "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/casual_not_formal/2478ece120c5bf5c87a4eabbc2d38ef04aec4242d60f451d3f775b33d6091fa5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "249438ad5c3ae59192b1e9d0cbb95c030e7bacf1d15e0ca7cbd3fb732b8431e2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eba073d3b6794ff5f88bd4bfb49380c421c792e8512414ccd26fc33868c1bb14", "generation_seconds_batch": 22.42272326897364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.823453426361084, "target_z": -0.8632811963626117, "ce": 6.038498878479004, "genuineness": 4.306513786315918, "identity": 0.3947490453720093, "hypothesis": "La porte est encore à faisneuve, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/249438ad5c3ae59192b1e9d0cbb95c030e7bacf1d15e0ca7cbd3fb732b8431e2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "24f9661334a9e85de318ecd2916dc15c0c43c692487886c59f042bb21464343b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d1da02de407a562293a5de093bfbe947528d7c2e8970615023f701f63fb4216", "generation_seconds_batch": 18.959574069012888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.573542356491089, "target_z": -0.31152343144762384, "ce": 5.990842819213867, "genuineness": 4.632867813110352, "identity": 0.8325330018997192, "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/casual_not_formal/24f9661334a9e85de318ecd2916dc15c0c43c692487886c59f042bb21464343b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "250f7e4fca9e7d2b74690ae285b3b825f9739f8ba7166c2ab17ca93f665ccaee", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a8ac26349be6739bb96d046923caa749ed8d24ee8eaef6b576a63aa28f858426", "generation_seconds_batch": 19.171465184073895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.609142303466797, "target_z": -0.9003906770462597, "ce": 5.6532793045043945, "genuineness": 4.13078498840332, "identity": 0.7081616520881653, "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/casual_not_formal/250f7e4fca9e7d2b74690ae285b3b825f9739f8ba7166c2ab17ca93f665ccaee.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": "253d913a9edb3bf41107bcf248e2c962bbc45197d10baf03b69476bcef1e7505", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "08a31ba41f8c6060122761d3e2e55caaa4d0b267b530cbac08c7c077813d4913", "generation_seconds_batch": 18.534652152913623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2620434761047363, "target_z": -0.6894531728259259, "ce": 6.117925643920898, "genuineness": 4.406930446624756, "identity": 0.5486727356910706, "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/casual_not_formal/253d913a9edb3bf41107bcf248e2c962bbc45197d10baf03b69476bcef1e7505.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "2563d01f06a403395a1fc75d8d98f26624612c54b5f4416d16de4288a555ac6a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "839a9d568564b650299e83e679d67e8c7e1b23da8d7d54a54e8381b7b566feb4", "generation_seconds_batch": 15.853106147027574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6644997596740723, "target_z": -0.140136784190801, "ce": 6.197352409362793, "genuineness": 4.557555198669434, "identity": 0.5792564153671265, "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/casual_not_formal/2563d01f06a403395a1fc75d8d98f26624612c54b5f4416d16de4288a555ac6a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "256532c31a2b178cc049c8e1a38deb81916d35d821d7c2169f8343f4cc01704e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2fe0853d2a909e20723a472c84cc3b0cfab116fba5a4ec23d4f6ee0ceb3dca3c", "generation_seconds_batch": 17.591778091038577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8483734130859375, "target_z": -0.7636718340004037, "ce": 6.006728172302246, "genuineness": 4.268857479095459, "identity": 0.7395377159118652, "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/casual_not_formal/256532c31a2b178cc049c8e1a38deb81916d35d821d7c2169f8343f4cc01704e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "259961d95bd147b7e99403627bf1a46b8836424bbeee6a1aae1e1fe761236da6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7650f82b577251e8d00f708def9473dacd18b80a542b7d5d01af6f123bc336d8", "generation_seconds_batch": 22.872305673081428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -0.9335937562940853, "ce": 6.109982967376709, "genuineness": 4.231201171875, "identity": 0.5896484851837158, "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/casual_not_formal/259961d95bd147b7e99403627bf1a46b8836424bbeee6a1aae1e1fe761236da6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "259ac3c3362dd3db03094f1dd4888eab1becf809da988ef496ab2913d1bc26c3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10e9f6666b36223374a89c04e7ae38f818728e1ef54dfdfc5c703035d043ddac", "generation_seconds_batch": 26.79759092105087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.324343204498291, "target_z": -0.2773437983937863, "ce": 5.597680568695068, "genuineness": 4.105680465698242, "identity": 0.39661115407943726, "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/casual_not_formal/259ac3c3362dd3db03094f1dd4888eab1becf809da988ef496ab2913d1bc26c3.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": "25f1ba1df491e3d981256255d4c7154f464bef86b22e7e2e97f49f805e696dd5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1468ffb728151fce4b24eedb3f486587e88e250eedfa99a18f70390bd28a13d4", "generation_seconds_batch": 21.88863990199752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.982940673828125, "target_z": -0.6562500337482723, "ce": 6.395919322967529, "genuineness": 4.093128204345703, "identity": 0.33731067180633545, "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/casual_not_formal/25f1ba1df491e3d981256255d4c7154f464bef86b22e7e2e97f49f805e696dd5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "25f8ed9de27fa8e23759f4319411f84ad969c05ce3f34cd5aa4d22bfdd7112d9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b111ff7a44cedb095549ade18a1b0ef04ac405e1860c00c83fc09c8e7b801819", "generation_seconds_batch": 17.073029723018408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.689108371734619, "target_z": -0.15588371463577955, "ce": 5.7128496170043945, "genuineness": 4.532451152801514, "identity": 0.47823020815849304, "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/casual_not_formal/25f8ed9de27fa8e23759f4319411f84ad969c05ce3f34cd5aa4d22bfdd7112d9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "2602847257bd3aa45240c75db6e98b104e3972a313ba2a2ef2107dce62fa481f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1eb6a28e05da68761f659c4c2867bd4cd73be491bcfb33a2ca7c420bbec7f29", "generation_seconds_batch": 26.15320115198847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.274503469467163, "target_z": -0.3867187877505064, "ce": 6.022613525390625, "genuineness": 4.231201171875, "identity": 0.504330039024353, "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/casual_not_formal/2602847257bd3aa45240c75db6e98b104e3972a313ba2a2ef2107dce62fa481f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "263cac238a293454d608d8ed8812ce2473d7f058deceb473c57591faef7412ed", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5118eca361c86fe1efef517828e4fdf415fbbc6721b4d858fbc58cbdb5d69e7c", "generation_seconds_batch": 22.600910209934227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5742545127868652, "target_z": -1.0195312772172769, "ce": 6.3641486167907715, "genuineness": 4.457139015197754, "identity": 0.6670429706573486, "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/casual_not_formal/263cac238a293454d608d8ed8812ce2473d7f058deceb473c57591faef7412ed.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": "26832309ef924e29c2f81235e1a26685e0f92e0ce7a0c7c4cc280fbf595be539", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "2461f4721ea3748d2bc36d62945995d7bb5b62e1a2519b2409acab64a854d27b", "generation_seconds_batch": 18.040472951019183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.953570604324341, "target_z": -0.047363314394794756, "ce": 6.141753673553467, "genuineness": 4.557555198669434, "identity": 0.7198333144187927, "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/casual_not_formal/26832309ef924e29c2f81235e1a26685e0f92e0ce7a0c7c4cc280fbf595be539.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "cb4857ce07ed6e805507fdecf14c75f7876fa2b58acf9bc7d6abf1a4a052e3bd", "generation_seconds_batch": 16.493021990987472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.224663734436035, "target_z": -0.6044922300779186, "ce": 6.53094482421875, "genuineness": 4.406930446624756, "identity": 0.33662277460098267, "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/casual_not_formal/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "26e430fe32f60836093b8606f761bbb99ad83c9f84d5923b015c927a032e09b3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5de3f37617f056e69382aa472cb6f768e9b0ae612f10d3c6195831bde3982e16", "generation_seconds_batch": 18.906176304910332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.461402654647827, "target_z": -0.30859378806865256, "ce": 6.1894097328186035, "genuineness": 4.582659721374512, "identity": 0.5195908546447754, "hypothesis": "Ik heb von die floatsies. Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in gueren.", "wer": 0.3333333333333333, "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/casual_not_formal/26e430fe32f60836093b8606f761bbb99ad83c9f84d5923b015c927a032e09b3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "26ffb346521df58ffa2c80b1710b87f7e82b708b006893da4df8723191220320", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5350f9644dfcb21f05573a8aafcb137f33dad049929a023afeb34aa3f283dfb8", "generation_seconds_batch": 23.360062603023835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.132460117340088, "target_z": -0.3222656235697445, "ce": 6.062326908111572, "genuineness": 4.250029563903809, "identity": 0.1191476508975029, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/26ffb346521df58ffa2c80b1710b87f7e82b708b006893da4df8723191220320.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "27768b44a4d5aa7af422401f22bdc66187251f01c1545dd704488e58d9610e01", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44aa3cf15eae53bdc6870ad3516625b474608855e822ec6e2596a444b7fee243", "generation_seconds_batch": 21.579793446930125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.457664728164673, "target_z": -0.45849609261107815, "ce": 6.0146708488464355, "genuineness": 4.444586753845215, "identity": 0.421610951423645, "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/casual_not_formal/27768b44a4d5aa7af422401f22bdc66187251f01c1545dd704488e58d9610e01.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "279dad62cfed96864993809b9c8a27f7de047c2446bbe926f6ddbb2dc81b6d76", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8cf7d43118d9207fc76671c716fb5339403fb5d2f381041f686eb2c12bd32e94", "generation_seconds_batch": 16.402832354069687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.6992187688294322, "ce": 6.316492557525635, "genuineness": 4.18099308013916, "identity": 0.6146308779716492, "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/casual_not_formal/279dad62cfed96864993809b9c8a27f7de047c2446bbe926f6ddbb2dc81b6d76.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "283a1f2b52a1d174e889930c8563c8cad7e964c36f3486fa87eab54908ca4a0b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1e68f335eaee0b7b5a76ed4512505356a3442d702cecc6121178685ee50924e", "generation_seconds_batch": 17.05932228907477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2154078483581543, "target_z": -1.3242188239619084, "ce": 6.252951145172119, "genuineness": 4.306513786315918, "identity": 0.5394591093063354, "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/casual_not_formal/283a1f2b52a1d174e889930c8563c8cad7e964c36f3486fa87eab54908ca4a0b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "287eaa8aa55f328b8b872b29972b8b7b31e6407c1d33755d1bd8e753da21eed0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48902b8b7218c5a373f1b32a28786d83d468f508a328f2c192d2f4f1676e5485", "generation_seconds_batch": 24.544297170010395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4476966857910156, "target_z": -0.22705083618278388, "ce": 5.839932441711426, "genuineness": 4.3881025314331055, "identity": 0.700701117515564, "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/casual_not_formal/287eaa8aa55f328b8b872b29972b8b7b31e6407c1d33755d1bd8e753da21eed0.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": "28b2dc2d5109e840eac6369124ee9b8a5a5b598820aea685e8bdd4b3ea8d03c3", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "288fc1f414a638b752049fd24a159026aa556c6b361ad5ce483889fac266c670", "generation_seconds_batch": 26.167363791028038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9163693189620972, "target_z": -1.5429687624841226, "ce": 5.998785495758057, "genuineness": 4.231201171875, "identity": 0.5410984754562378, "hypothesis": "Ich schrieb die Nummer auf den Umschlag, lie sie genau, und prüfe die letzten beiden Ziffern. Ruf an, wenn du Zeit hast.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/28b2dc2d5109e840eac6369124ee9b8a5a5b598820aea685e8bdd4b3ea8d03c3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "28bef6af9c9637dd422c3fa758390ed25b5dc728205f35a763cb20eaf58e7877", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28eb0aabeed4e2972e1e43963383d2e75c87b93daeb8a46a96556a26b8d44b50", "generation_seconds_batch": 26.823047554004006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.2265625413533585, "ce": 6.340320587158203, "genuineness": 4.105680465698242, "identity": 0.6237126588821411, "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/casual_not_formal/28bef6af9c9637dd422c3fa758390ed25b5dc728205f35a763cb20eaf58e7877.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "28c6a54c492ddc0bff2dd9625e0bd75ee273330dd65d5f975ffa380ef99b85c1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3767b68476896b3de2a906b45d81eb9bc68fcfba6b5a61a53775d432ae4b6def", "generation_seconds_batch": 17.349169188993983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.726332426071167, "target_z": -0.10174556082778452, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.5948123931884766, "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/casual_not_formal/28c6a54c492ddc0bff2dd9625e0bd75ee273330dd65d5f975ffa380ef99b85c1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "2911c912e91303591f664b03cff180d89bb43570007d0a36151df55bbeb99d43", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc30578f2df7f5e304f70a0c579d73a4be76d4015e9bc7d782e50734633eec00", "generation_seconds_batch": 22.075816667987965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8046741485595703, "target_z": -0.10473637088700483, "ce": 5.808161735534668, "genuineness": 4.1998209953308105, "identity": 0.5322812795639038, "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/casual_not_formal/2911c912e91303591f664b03cff180d89bb43570007d0a36151df55bbeb99d43.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": "292859e66853ac51bd48ca5c55eb5ab45903212d45c75763fa736490a8231114", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "195499b6de055fb05fc383134c018ddd0d04019acb122ebe8ed812224ce93930", "generation_seconds_batch": 22.76763163297437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2346317768096924, "target_z": -0.6005859468658923, "ce": 6.157639026641846, "genuineness": 4.344170093536377, "identity": 0.5809832811355591, "hypothesis": "Las llaves están en la mesa de la cocina, mire en el cajón y luego junto a la ventana. Llévatelas cuando te vaya.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/292859e66853ac51bd48ca5c55eb5ab45903212d45c75763fa736490a8231114.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "2930bca94a3d8386ce77ca0f0316b656484d4fa82b3af365da2a5cb7e95ca330", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86f550088aecb69158a0ad5f7d9ef42af29186924f22f2f0096c2969cd5734b9", "generation_seconds_batch": 23.34196627791971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.022812604904175, "target_z": -0.4863281031683918, "ce": 6.229123115539551, "genuineness": 4.268857479095459, "identity": 0.1273660957813263, "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/casual_not_formal/2930bca94a3d8386ce77ca0f0316b656484d4fa82b3af365da2a5cb7e95ca330.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "2969d4962131d23b9b8b0560c2e17da5eaf1689619ccefbc29cba6cc0c48b079", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "342163f3fff01b2dd0a520917dc4b4b3d993fed4083d49078cd4638351aeb1ae", "generation_seconds_batch": 23.2453904079739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9660305976867676, "target_z": 0.023681631352249398, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.3543730080127716, "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/casual_not_formal/2969d4962131d23b9b8b0560c2e17da5eaf1689619ccefbc29cba6cc0c48b079.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "29796548acff61a4d5d876231fca0bc70c985cbb911b3c4148195f37fc0064de", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e416df7c12eaa74eaac7b4033919d0c463c91fdf92c0025835fafdfa62db0617", "generation_seconds_batch": 17.26383608893957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4314987659454346, "target_z": -0.23193363606039735, "ce": 6.260893821716309, "genuineness": 4.432034492492676, "identity": 0.2986259460449219, "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/casual_not_formal/29796548acff61a4d5d876231fca0bc70c985cbb911b3c4148195f37fc0064de.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "297c1b47b00a375e4cce73f732e112c75b7d8f48e9ce78b34dffee4a4e94a73c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "23fc9c5a656e91c5b0d22260e152d4ce6f3991a7c6883e459f948e97369cecbb", "generation_seconds_batch": 23.913284947979264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6246280670166016, "target_z": -0.3442382439545025, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.7409967184066772, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/297c1b47b00a375e4cce73f732e112c75b7d8f48e9ce78b34dffee4a4e94a73c.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": "29f727f9aa269f5750c607dd122098061b800119a6d5c75461b36a16ab96afc9", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "c26c5fcadbdfaf60286cfeb43d75e3f713322516672ec781019e6536f2b1478d", "generation_seconds_batch": 17.15244798909407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8918938636779785, "target_z": 0.00610350202856659, "ce": 6.141753673553467, "genuineness": 4.218649387359619, "identity": 0.38308030366897583, "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/casual_not_formal/29f727f9aa269f5750c607dd122098061b800119a6d5c75461b36a16ab96afc9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "2a013f7b66e6b8285abf51998a83cea2a3e28d42bef48723ef915247f70560aa", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "89489e6b7b425d9f21f19a9452ca8dc6fd67064108224fe4d0a6789e09830afa", "generation_seconds_batch": 24.85562670393847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.358519077301025, "target_z": 0.4101563101747878, "ce": 6.086154937744141, "genuineness": 4.218649387359619, "identity": 0.6563575267791748, "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/casual_not_formal/2a013f7b66e6b8285abf51998a83cea2a3e28d42bef48723ef915247f70560aa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "2a188a4a700a7f3c5d1e88c090d35e20ae888d1cbcd52e8908af5fcdd17025d8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c43efe9e920c69849512c7d2c5f4dd18e91ab454d88f02607fddebc4b0c51ec", "generation_seconds_batch": 16.30936093092896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.840808153152466, "target_z": -0.15795898753604853, "ce": 6.324435234069824, "genuineness": 4.375550270080566, "identity": 0.43727582693099976, "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/casual_not_formal/2a188a4a700a7f3c5d1e88c090d35e20ae888d1cbcd52e8908af5fcdd17025d8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "2a92c22242b1deadb503af051224cb4966057682342e44df28af7c099575e974", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1842e0885e30bdcda9ec4581ca182a69de51d12708056c1e2ff29d0cace3080", "generation_seconds_batch": 16.4145659430651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.7343749932064861, "ce": 6.229123115539551, "genuineness": 4.105680465698242, "identity": 0.42279380559921265, "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/casual_not_formal/2a92c22242b1deadb503af051224cb4966057682342e44df28af7c099575e974.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "2afb719fc2d9e07de42be8a1d4604f3bb34f205e0187a7b8e506043848f5488c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff3222df578812c0d7dec5612711dea24b63c1acbd70d13a28b21dbaec1c54d3", "generation_seconds_batch": 17.03976545901969, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.128906170794088, "ce": 6.173524379730225, "genuineness": 3.8420867919921875, "identity": 0.30045393109321594, "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/casual_not_formal/2afb719fc2d9e07de42be8a1d4604f3bb34f205e0187a7b8e506043848f5488c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "2b8c7451df07d484ed646fda56c6744d5626ae0d243f164761ae9fa3734ee688", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fabac03a1bc4dbbf0cfecb6bf9a3a776d2754a433cfb03bd0a245efcddbe5fc3", "generation_seconds_batch": 26.885386556969024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.461402654647827, "target_z": -0.04003909525781797, "ce": 5.982900142669678, "genuineness": 4.2374773025512695, "identity": 0.3308630585670471, "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/casual_not_formal/2b8c7451df07d484ed646fda56c6744d5626ae0d243f164761ae9fa3734ee688.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "2bb3afb833a6fb7850b9e1a02bf2b5a911d2315ff38380369abe5780f195da0a", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1dfba3173994d48a2250bf72f60f307b16021f888ff0be2cf014c20f79fe9586", "generation_seconds_batch": 17.143384692026302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3460588455200195, "target_z": 0.32080071032448443, "ce": 6.197352409362793, "genuineness": 4.582659721374512, "identity": 0.6369290351867676, "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/casual_not_formal/2bb3afb833a6fb7850b9e1a02bf2b5a911d2315ff38380369abe5780f195da0a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "2be0303ce99a02dbcffd97c166d542d433732dd581d6c7b8467049f497f0c0ec", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7330e0779a4c8683fc20a70070642a1466985989f05ac8504c5f9de04d97c3e", "generation_seconds_batch": 16.529907124931924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2253761291503906, "target_z": -1.0937500193608733, "ce": 6.117925643920898, "genuineness": 4.306513786315918, "identity": 0.7778700590133667, "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/casual_not_formal/2be0303ce99a02dbcffd97c166d542d433732dd581d6c7b8467049f497f0c0ec.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "2bf278cd6e416a0bde558f2849a2f4d5e6fb5b3c95a63dfca395eb5229b968fa", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4890170a997a5c68a4ec26db1adf018fab1bbed7986f360b4f68c602bfd1ea9d", "generation_seconds_batch": 21.67579561995808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0078606605529785, "target_z": -0.7988281473718557, "ce": 6.062326908111572, "genuineness": 4.419482707977295, "identity": 0.6174840927124023, "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/casual_not_formal/2bf278cd6e416a0bde558f2849a2f4d5e6fb5b3c95a63dfca395eb5229b968fa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "2cdbcd3620547f22cb0bc45410722f08b019d0000328d1c477393ed5d1076224", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4d7eeb36f2d1af5ab6d0ef2fc3a89a22dfbb40d484ed83c3c3a1473453c6c08", "generation_seconds_batch": 22.08326040999964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5774586200714111, "target_z": -1.8515626050977807, "ce": 6.324435234069824, "genuineness": 4.055472373962402, "identity": 0.5396100282669067, "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/casual_not_formal/2cdbcd3620547f22cb0bc45410722f08b019d0000328d1c477393ed5d1076224.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "2ce17cac0006516a7d048cfc18158749153b0366bae1f5251a5aa617e670cc0a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "660969f16b20f3a756c0e428ffc2bbf4f8922e49611b516a5c854b9e28b4c38e", "generation_seconds_batch": 23.456846726941876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.506258487701416, "target_z": -0.2890625375364082, "ce": 5.967014789581299, "genuineness": 4.7458367347717285, "identity": 0.774105966091156, "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/casual_not_formal/2ce17cac0006516a7d048cfc18158749153b0366bae1f5251a5aa617e670cc0a.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": "2d00feb526158635acacad6423bcfcd52779494c953a98c72b4730ef1fbf3f96", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0447f00a44fcb170eaea6251282c2d2a36920ca086cfe128c302d6b7530dab1d", "generation_seconds_batch": 17.25393615302164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9946885108947754, "target_z": -0.07128903165986147, "ce": 6.197352409362793, "genuineness": 4.65797233581543, "identity": 0.6377850770950317, "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/casual_not_formal/2d00feb526158635acacad6423bcfcd52779494c953a98c72b4730ef1fbf3f96.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "2d7140860705bc05198c9b3f23fddad7b6b00e30b818cf1e3465170efe333fc9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "deef0c5d3bc0328408083dc9eead243c2e1d26326b7b7de09ae814fd52a298ce", "generation_seconds_batch": 26.124393612961285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.044528007507324, "target_z": 0.01757805522881177, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.6831709742546082, "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/casual_not_formal/2d7140860705bc05198c9b3f23fddad7b6b00e30b818cf1e3465170efe333fc9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "2d8fe81ebfe311472a57982125420ab03f39f99006762ab5ea7ccf14f7cfb0f1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e4913dcaa53e41c07382bd9b53c2405f6dc6a9f44b22b3ad374621f10901d2c", "generation_seconds_batch": 21.978963394998573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.364215135574341, "target_z": -0.2138672045590203, "ce": 6.109982967376709, "genuineness": 3.9299514293670654, "identity": 0.2719464898109436, "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/casual_not_formal/2d8fe81ebfe311472a57982125420ab03f39f99006762ab5ea7ccf14f7cfb0f1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "2da573f4d524abec32896ebb8e41cd9f686f1c8bf38ecd68d62bb80b8860b57c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62425a72aee4a4590e94fc8a189594124f7103fcec2b7693788421ad791a10b0", "generation_seconds_batch": 17.670204249909148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.7968750239037565, "ce": 6.157639026641846, "genuineness": 4.3881025314331055, "identity": 0.634348452091217, "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/casual_not_formal/2da573f4d524abec32896ebb8e41cd9f686f1c8bf38ecd68d62bb80b8860b57c.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": "2e24803d47508ec7754bd4efde6b605f91f74496ce45df8743df944ea2264bad", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b36c1a18e85b54eafc756ffe40e34db2707e4fdcefb60be566b4560ec894c7d", "generation_seconds_batch": 17.47923596005421, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.6640625077843678, "ce": 5.967014789581299, "genuineness": 4.494794845581055, "identity": 0.5986142754554749, "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/casual_not_formal/2e24803d47508ec7754bd4efde6b605f91f74496ce45df8743df944ea2264bad.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": "2e6c444934af19f3b279063c2bd519c20a64015a7e4b2fc717411e63699f4428", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "231856edf65bc432ebe3d57fa05307bf9222d3fb04b7591a03592c1971fcb778", "generation_seconds_batch": 16.770892351982184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.441466808319092, "target_z": -0.43554688282838877, "ce": 5.879645347595215, "genuineness": 4.344170093536377, "identity": 0.7466495037078857, "hypothesis": "Quelqu'un avait placé la petite chaise bleue. Elle était ici, près de la grande bibliothèque. Nous la remettrons après le déjeuner.", "wer": 0.09523809523809523, "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/casual_not_formal/2e6c444934af19f3b279063c2bd519c20a64015a7e4b2fc717411e63699f4428.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "2e9b51b039659e7af2c917a203b96c38a5a0802376eff41051cbbc069fe7833a", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b739afbab774b0332bb17dcc74dc8bf6abda65ccc6ecc6187711a6f8c864d353", "generation_seconds_batch": 16.821575233945623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.813485622406006, "target_z": -0.9101561795533037, "ce": 6.268836498260498, "genuineness": 4.413206577301025, "identity": 0.7125831842422485, "hypothesis": "La porte est encore, est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/2e9b51b039659e7af2c917a203b96c38a5a0802376eff41051cbbc069fe7833a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "2f47cafdac8119a109a0116dd81a74a7a1342b83513a2fe1a25a247c11c47cbc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9125692ec3b6775f336f28a356c6779f737ead8266aed2f03b29d6bff9f33b5", "generation_seconds_batch": 22.42272326897364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.8867187681443183, "ce": 6.229123115539551, "genuineness": 4.570107460021973, "identity": 0.7181476354598999, "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/casual_not_formal/2f47cafdac8119a109a0116dd81a74a7a1342b83513a2fe1a25a247c11c47cbc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "300afc65362dac880b04e8a639d617bdbb79df2f61036b16b6ea0909cd33cc75", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09c825ee86c1e8e76515bbcba358000671adcdf23192d8f1bea0dcc02f3cbc37", "generation_seconds_batch": 23.399708233075216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2154078483581543, "target_z": -1.332031288864219, "ce": 6.070269584655762, "genuineness": 4.193545341491699, "identity": 0.5552060008049011, "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/casual_not_formal/300afc65362dac880b04e8a639d617bdbb79df2f61036b16b6ea0909cd33cc75.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": "31215cb252bdc63aa7b9210baab40f5aef8cfaf670046ef1868f9c67aa06c6ea", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9812e46cb5d769bcff4f36eff51608b5265603c92bb404314eeda08a58461af9", "generation_seconds_batch": 16.53924127900973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6839020252227783, "target_z": -1.0429688419525531, "ce": 6.0146708488464355, "genuineness": 4.143336772918701, "identity": 0.5284426212310791, "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/casual_not_formal/31215cb252bdc63aa7b9210baab40f5aef8cfaf670046ef1868f9c67aa06c6ea.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": "318f5f2f5c1c30ff351ca2f7cfa6f1326152096f0b9213f3656723a5303a0aba", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "13844bf68d3ec6b55a6857439f42515ca7196b8d3a65e894381ec0bf1d7decae", "generation_seconds_batch": 17.02780994400382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.604158401489258, "target_z": -1.0429687312970257, "ce": 6.062326908111572, "genuineness": 4.162164688110352, "identity": 0.5800870656967163, "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/casual_not_formal/318f5f2f5c1c30ff351ca2f7cfa6f1326152096f0b9213f3656723a5303a0aba.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "31a4aebb4c78f3203763d5b6e392b86076d32a12a5feae743a6cb0a864cf9be3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "acfab6ad24bc744b069b47233408fe6c3ced834387e0b64f7fe136ccd4530d88", "generation_seconds_batch": 19.15707281907089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1798079013824463, "target_z": -0.520507847759364, "ce": 5.990842819213867, "genuineness": 4.494794845581055, "identity": 0.7150969505310059, "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/casual_not_formal/31a4aebb4c78f3203763d5b6e392b86076d32a12a5feae743a6cb0a864cf9be3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "31f078d0c864a80c379f745f4ab3a9835218b9a37446e1874c32dcc8746e4ffd", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8c0059211a35217a53009034ed795e8941267285f1f696c963608e9611b62a5", "generation_seconds_batch": 23.1512352040736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017116546630859, "target_z": 0.17578132574392535, "ce": 6.1020402908325195, "genuineness": 4.13078498840332, "identity": 0.44542163610458374, "hypothesis": "I found little box. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/31f078d0c864a80c379f745f4ab3a9835218b9a37446e1874c32dcc8746e4ffd.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "321a43e5abc78d30bea2cf700113d363289abddebe032c8422e6947e3c2d136e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2dbe610e12eb0ebc77e70f4882abb29c0caf751ef97a1026cc9eeced35e7247f", "generation_seconds_batch": 17.38358975807205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8433892726898193, "target_z": -0.8359375611144173, "ce": 6.157639026641846, "genuineness": 4.394378662109375, "identity": 0.7680085897445679, "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/casual_not_formal/321a43e5abc78d30bea2cf700113d363289abddebe032c8422e6947e3c2d136e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "32352a37b2e08edd3badcea3dfa73c922e84ebd601f88b7534545e5fe8dea33f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2d79f4f36e565173947504077de3b5ef84288855942c168375d0333bb2d81f4", "generation_seconds_batch": 15.27358686097432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.48881459236145, "target_z": -0.12670897942142947, "ce": 6.054384231567383, "genuineness": 4.344170093536377, "identity": 0.6919535994529724, "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/casual_not_formal/32352a37b2e08edd3badcea3dfa73c922e84ebd601f88b7534545e5fe8dea33f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "3290ead63d18e9a911c892e397c0bc0bd2d7c30125c77447ec6f2a6da608b2c9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "986262698b02c6a345adda8fa8e8c8d8c58ddcd36f984de4fa3aa7e2d004f2e0", "generation_seconds_batch": 17.655254842014983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.413342475891113, "target_z": 0.5195311552085425, "ce": 6.125868320465088, "genuineness": 4.2374773025512695, "identity": 0.42735883593559265, "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/casual_not_formal/3290ead63d18e9a911c892e397c0bc0bd2d7c30125c77447ec6f2a6da608b2c9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "32981ab324649de1bc36d31268193c79b39342385033865afa23a63f73803941", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a7b56a06743819a5befcaa00ba93339f75bcf29762696c8795c59ba7d7e70e07", "generation_seconds_batch": 25.458952556014992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9622926712036133, "target_z": 0.22070315545436203, "ce": 5.796247482299805, "genuineness": 4.218649387359619, "identity": 0.6456866264343262, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/32981ab324649de1bc36d31268193c79b39342385033865afa23a63f73803941.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": "33a57d23ef1bd84bc704b8921ac9002358bba1f214bea52af4e3c5e38e298d75", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "20829efead5ee92fab882b0c4c21eb4d89836b85321edcace64efc4cae7ca615", "generation_seconds_batch": 26.137216016068123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301915407180786, "target_z": -0.5937499543774857, "ce": 6.23706579208374, "genuineness": 4.381826400756836, "identity": 0.7252928614616394, "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/casual_not_formal/33a57d23ef1bd84bc704b8921ac9002358bba1f214bea52af4e3c5e38e298d75.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "33dc1c4297cd4f2e424377d2f50f7829b9574d85ebb1a55a9e0e664b71a6885e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fdaa9c47da617a79785a60d9cbcc3e8a5005689b44f57ccf78d862179e6ff5c", "generation_seconds_batch": 16.571312663028948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4977149963378906, "target_z": -1.8906250554013984, "ce": 6.2767791748046875, "genuineness": 4.218649387359619, "identity": 0.4498958885669708, "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/casual_not_formal/33dc1c4297cd4f2e424377d2f50f7829b9574d85ebb1a55a9e0e664b71a6885e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "33de733e019a525bd7939d4449a43da19d8cb968f8c1b08c0d63edcb543f13a1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "40a973e1cd42d3e28f6e1bde4d80ad81aa0f5699724539a5d196ffd4da8ec694", "generation_seconds_batch": 21.910004258039407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.185504198074341, "target_z": -1.3945312099496394, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.6542216539382935, "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/casual_not_formal/33de733e019a525bd7939d4449a43da19d8cb968f8c1b08c0d63edcb543f13a1.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "5df27113edd504eb482dab9f9347ed706fa8ecb0be95a8b2826eee2f67eab58c", "generation_seconds_batch": 17.21664079092443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7130937576293945, "target_z": -0.13964836898378924, "ce": 5.9431867599487305, "genuineness": 4.432034492492676, "identity": 0.6275526881217957, "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/casual_not_formal/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "340c19908e46124697ac547a7b4478f14e69094c89ea34f7047912e001779d78", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c5c595cd02b6749795f5b305317db595616839320e6ce0287374ac035183aba", "generation_seconds_batch": 23.49804195703473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.306899309158325, "target_z": -0.42578127097041446, "ce": 6.117925643920898, "genuineness": 4.356721878051758, "identity": 0.5738804340362549, "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/casual_not_formal/340c19908e46124697ac547a7b4478f14e69094c89ea34f7047912e001779d78.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "3413ce5d93f84fb32c05cbd434514422702a32feda1a95f7877e8f2bd1e732a3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af08f600e2bbfb67ba4fcfd8943ed783ef016f497128127dd2efc378ca83b9d3", "generation_seconds_batch": 16.250904027954675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.818469524383545, "target_z": -0.8613281770254486, "ce": 5.967014789581299, "genuineness": 4.218649387359619, "identity": 0.6096285581588745, "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/casual_not_formal/3413ce5d93f84fb32c05cbd434514422702a32feda1a95f7877e8f2bd1e732a3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "346a06100e5f237ab94c90cd8878c0e03e5ff5b8b3e4af68a127617c9cb28110", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d97af95bf69fe87bc0ec6a8f32c5b69c59a0016deed907a122d52cff4e021184", "generation_seconds_batch": 23.49804195703473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8583412170410156, "target_z": -0.7500000039802257, "ce": 6.165581703186035, "genuineness": 4.394378662109375, "identity": 0.5707445740699768, "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/casual_not_formal/346a06100e5f237ab94c90cd8878c0e03e5ff5b8b3e4af68a127617c9cb28110.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "34773dffc6f58ee21d720502d77ddfe9ad25ba7c7e40e1e3a18db29bc7cc4b90", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59a9a81d59d98e005b601bbf37f86938e9c531d3c6f0882bb4b33eb3aec4dbbc", "generation_seconds_batch": 20.2761810059892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1449201107025146, "target_z": -0.48925778921812657, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.7744385600090027, "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/casual_not_formal/34773dffc6f58ee21d720502d77ddfe9ad25ba7c7e40e1e3a18db29bc7cc4b90.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "73c6548d2e68620f91f8e2247ef2dd5eda4c7dbf7179d9dc97770210abfd99f8", "generation_seconds_batch": 17.52830210106913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.677138328552246, "target_z": -1.6484375809811231, "ce": 6.459460258483887, "genuineness": 4.2249250411987305, "identity": 0.4150580167770386, "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/casual_not_formal/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "349457c98b4035bf27cadaf391a4c6f6da0c9e99e7fc474e009d8ca91215c7d4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "476455de04bbcf2b8269fa4c6484ed00efa9aedb703e1638cb84db65d66e3d26", "generation_seconds_batch": 21.522072001942433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3118834495544434, "target_z": -0.34179684650293507, "ce": 5.506340026855469, "genuineness": 3.873466968536377, "identity": 0.2839950919151306, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás libre.", "wer": 0.06666666666666667, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/349457c98b4035bf27cadaf391a4c6f6da0c9e99e7fc474e009d8ca91215c7d4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "34958e66c76a3d5ab33359ad7af2d06efefa092d180ad28cccca4d687ddf26c9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2828a01baacf327a9aa3a538d58dd664c0511ba8c6e7f353dbf4b13e6e95b4e", "generation_seconds_batch": 27.497890384984203, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0160489082336426, "target_z": -1.5546873974633946, "ce": 6.38003396987915, "genuineness": 4.337893962860107, "identity": 0.4875556230545044, "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/casual_not_formal/34958e66c76a3d5ab33359ad7af2d06efefa092d180ad28cccca4d687ddf26c9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "349f4664e64838c1f266f9e07557486d82bf464febc25f9efa9f7875de1fe822", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1ac2765ce121fa6b848fb25a1cd0022be6dfe14280fa644b9a932202aca24a8d", "generation_seconds_batch": 23.17944325506687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.120000123977661, "target_z": -0.5341797199284668, "ce": 6.24500846862793, "genuineness": 4.1998209953308105, "identity": 0.27926647663116455, "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/casual_not_formal/349f4664e64838c1f266f9e07557486d82bf464febc25f9efa9f7875de1fe822.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "3579d867b6b353058110b16ea895da04db33c2c1e68917ea22154d5eac283246", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fea8f1d91ab7c43fa75dd957f67e7a8f1babce52157043855401ff307973270f", "generation_seconds_batch": 21.420616014976986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.984374934094588, "ce": 6.1894097328186035, "genuineness": 4.344170093536377, "identity": 0.6224848031997681, "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/casual_not_formal/3579d867b6b353058110b16ea895da04db33c2c1e68917ea22154d5eac283246.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "357c9a9a2622b4b493899758899d37faa04ec13781cc8276942af6b905b939ea", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df1c8cbf8838b1f6c473c2f1786029845b4b15442b28c6cf46c732c5baa2f25b", "generation_seconds_batch": 17.831452472950332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.7929687697914722, "ce": 6.078212261199951, "genuineness": 4.482243061065674, "identity": 0.5966481566429138, "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/casual_not_formal/357c9a9a2622b4b493899758899d37faa04ec13781cc8276942af6b905b939ea.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": "359bc21c3dfbc1ae389b704ee3910eb00ede89f607ee1353cbcf3ebd22e6194a", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d13f14105c8b93347cebd8708eafe7d416e7c9985bd7b93da68a65c15b914202", "generation_seconds_batch": 16.54173002706375, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.276283264160156, "target_z": 0.29980468020225454, "ce": 6.165581703186035, "genuineness": 4.218649387359619, "identity": 0.651786744594574, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Wanneer we kunnen praten.", "wer": 0.08695652173913043, "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/casual_not_formal/359bc21c3dfbc1ae389b704ee3910eb00ede89f607ee1353cbcf3ebd22e6194a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "35d110d6680e49227e97c00fc4229635fc4246ebb187dc8276ddeeb1c817047f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57a968b99de72bffbe1662aab5266eba6d4f894c0cfab0f04858687a58afd3a4", "generation_seconds_batch": 15.853106147027574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7088217735290527, "target_z": -0.7070312683772269, "ce": 5.863759994506836, "genuineness": 4.293961524963379, "identity": 0.6119124293327332, "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/casual_not_formal/35d110d6680e49227e97c00fc4229635fc4246ebb187dc8276ddeeb1c817047f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "35e8432f240e4ea4acc87233a60d6c33330fd98f48c0512148c7fd8bdc7e2ed8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "377653698f9f3bb2e37719bbaf075d359bc704aae8d68689a0dfa1ba838d50c1", "generation_seconds_batch": 23.06769301602617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.4453125195449599, "ce": 5.967014789581299, "genuineness": 3.9550557136535645, "identity": 0.6753238439559937, "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/casual_not_formal/35e8432f240e4ea4acc87233a60d6c33330fd98f48c0512148c7fd8bdc7e2ed8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "363290523fb51c48d66a0d0f38ed4956e29fb4a05e44e575e202399112050bb7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "29e63244d173c5b69dae935d17778a2a932f552a1c06be0ce7c6ebaf12e0cc2c", "generation_seconds_batch": 22.776833641924895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.813485622406006, "target_z": -0.9531249729636038, "ce": 6.109982967376709, "genuineness": 4.432034492492676, "identity": 0.7106737494468689, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/363290523fb51c48d66a0d0f38ed4956e29fb4a05e44e575e202399112050bb7.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": "3641f9b19a70ec04179e03c6de58e03d8411e059b0c7381292c30dda05c1e98e", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "377b9450b560008325e2bb4d6481e1ae3d031eedd5fb762ad7c5ab11ebb2203b", "generation_seconds_batch": 21.793964514043182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2171878814697266, "target_z": -0.5800780088075332, "ce": 6.499174118041992, "genuineness": 4.174716949462891, "identity": 0.3978056311607361, "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/casual_not_formal/3641f9b19a70ec04179e03c6de58e03d8411e059b0c7381292c30dda05c1e98e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "3663034e01d08306799233152fb0642e45ffafbbde9ebad67e16a9210a852053", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cb5916aa5457bcb15662c2e9507a08c75de54797ec95db06a02bef9f29a0f5e", "generation_seconds_batch": 22.495720722014084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7536778450012207, "target_z": -0.6953124484275566, "ce": 6.213237762451172, "genuineness": 4.20609712600708, "identity": 0.3853604793548584, "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/casual_not_formal/3663034e01d08306799233152fb0642e45ffafbbde9ebad67e16a9210a852053.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": "368545aef7ca4987f130f06063b48dac6143ee231e142925915fab4674f575a5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 129, "seconds_generated": 10.32, "mp3_sha256": "ad06f179ea80f66b7a7414cddaf16bcf2013eb518ce50ed2546718c97fd782e4", "generation_seconds_batch": 18.93227247998584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8034281730651855, "target_z": -0.13842773866816963, "ce": 6.332377910614014, "genuineness": 4.256305694580078, "identity": 0.2738149166107178, "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": 0.120000000000001, "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/casual_not_formal/368545aef7ca4987f130f06063b48dac6143ee231e142925915fab4674f575a5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "36aa48cda6ad2b9e034b4fe7d08a45cf9bda8f889fae9f1af76a26cc2690b0bc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "375aa14afd036e463d4011e57b256aa6a84bfa8f39d558d10a4475bc21ce56a3", "generation_seconds_batch": 23.81996055901982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8932290077209473, "target_z": -0.7207031048571682, "ce": 5.792276382446289, "genuineness": 4.394378662109375, "identity": 0.6366649866104126, "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/casual_not_formal/36aa48cda6ad2b9e034b4fe7d08a45cf9bda8f889fae9f1af76a26cc2690b0bc.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": "36bcdc9534932840fa02a05ef1802da086c2a811903e610770b9c0e3e7e5cc2c", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9d017cefb50bf4ac2d305f3e78b038b1f14f85a1d562987a0dceb9aed09042f7", "generation_seconds_batch": 24.710422587930225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4589107036590576, "target_z": -0.49316408828462077, "ce": 6.221180438995361, "genuineness": 4.325342178344727, "identity": 0.5361361503601074, "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/casual_not_formal/36bcdc9534932840fa02a05ef1802da086c2a811903e610770b9c0e3e7e5cc2c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "36d15c289417dfd67dc1c6484efe4a570d093d5fb23be5e03905a3fb6a49dd20", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28f11d6bbb991585963319001590df45581d3bee02b0a26b050853872ac173d2", "generation_seconds_batch": 16.477766836993396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0128445625305176, "target_z": -0.812500100510077, "ce": 6.157639026641846, "genuineness": 4.469690799713135, "identity": 0.6664996147155762, "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/casual_not_formal/36d15c289417dfd67dc1c6484efe4a570d093d5fb23be5e03905a3fb6a49dd20.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "36dbdbc75a2fd241777c30edaa6627dbc498ef826e343f168969c898f1881113", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1d58e0d71e9fec56ac140d759b2d8253bfb09c70b38ba236ba94cccd4821b5ac", "generation_seconds_batch": 21.86924408201594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.252075672149658, "target_z": -0.5625000434855338, "ce": 6.23706579208374, "genuineness": 4.68307638168335, "identity": 0.6149165630340576, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/36dbdbc75a2fd241777c30edaa6627dbc498ef826e343f168969c898f1881113.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "374dfe34753eec7e92424877e962b137aa8e6bf4b4b1f19b80324c91d23e5e88", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cab536a3c8dc4881121c040373abc0c795ed65a5820f5d29d146862b653413be", "generation_seconds_batch": 22.762878520064987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5685582160949707, "target_z": -0.12817382380031833, "ce": 6.070269584655762, "genuineness": 4.005263805389404, "identity": 0.350780725479126, "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/casual_not_formal/374dfe34753eec7e92424877e962b137aa8e6bf4b4b1f19b80324c91d23e5e88.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "378045692050bf091130d34e91a8ab7817079addc0483d6ebe451d676aeaa036", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "791e6def44f041dc0b3ad83027e3fe3428710439196eec064d18d6404bcc4cd5", "generation_seconds_batch": 22.988013435038738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.506258487701416, "target_z": -0.20556644944360158, "ce": 6.1020402908325195, "genuineness": 4.20609712600708, "identity": 0.6452693939208984, "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/casual_not_formal/378045692050bf091130d34e91a8ab7817079addc0483d6ebe451d676aeaa036.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "380edb8802dd7a1bcd42d26507d34a4b7140a8f37bb8af4a9a80bed7e31e33cb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98dd2788279136d86cdbf76d4c22d35b45dc2d3db4afd70fab7e54862f1ea3b4", "generation_seconds_batch": 22.623998783063143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4165468215942383, "target_z": -0.3535156677572022, "ce": 5.855817794799805, "genuineness": 4.457139015197754, "identity": 0.6784250140190125, "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/casual_not_formal/380edb8802dd7a1bcd42d26507d34a4b7140a8f37bb8af4a9a80bed7e31e33cb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "38bfd3201ab924ab95cb40407ab29eb82574451714c3659b0242cc050fa27cc0", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "19d4e1fe69f8448bfa4eb7250fb8bc6bb17f20c888d9752d0be6c42ee10cbfbe", "generation_seconds_batch": 16.112435367074795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.7402343578025149, "ce": 6.2767791748046875, "genuineness": 4.268857479095459, "identity": 0.4813668727874756, "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/casual_not_formal/38bfd3201ab924ab95cb40407ab29eb82574451714c3659b0242cc050fa27cc0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "38ff5525b0960f670ce4fde3cdc6afb29e53c858861c90ca1c34ab5e8229b1e3", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36149e80e73dacdb70f4c0103ffc5d70f38a0c5f44cf3ffa0de644b42ef4029b", "generation_seconds_batch": 16.862784418975934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7387256622314453, "target_z": -0.6425781967464922, "ce": 6.340320587158203, "genuineness": 4.381826400756836, "identity": 0.28044041991233826, "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/casual_not_formal/38ff5525b0960f670ce4fde3cdc6afb29e53c858861c90ca1c34ab5e8229b1e3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "392054a4f640b01a4340f09f0f0b51231b532dcdda6b194f5c2fb0336345a27d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f86afb1393a363be65f639eb70f791325b293423a44eb0a0410d99bd1927aae0", "generation_seconds_batch": 24.95866973802913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.017116546630859, "target_z": 0.12548833170973456, "ce": 6.260893821716309, "genuineness": 4.758388519287109, "identity": 0.6563143730163574, "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/casual_not_formal/392054a4f640b01a4340f09f0f0b51231b532dcdda6b194f5c2fb0336345a27d.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": "393ff30239a0bed716f7cb595d5a162eb981823cf4efc5286bd590438b0fa0bb", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ef672992eb6ff965e2cf1ca674745ffffff3042d93189f8d609019a74d68add", "generation_seconds_batch": 23.223817850928754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9622926712036133, "target_z": -0.1982421369568289, "ce": 6.229123115539551, "genuineness": 4.720732688903809, "identity": 0.5659294128417969, "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/casual_not_formal/393ff30239a0bed716f7cb595d5a162eb981823cf4efc5286bd590438b0fa0bb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "3a0c44ec7489ca3539a97b2faae1d44724363ce33950852dd59f066ccf7eba0e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bc77ae8dac0d6408703b33c9327f35b2d5e2df01c5e3ed6df5a534780175dc8", "generation_seconds_batch": 22.112723449943587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7038378715515137, "target_z": -0.8867188264734582, "ce": 6.062326908111572, "genuineness": 4.13078498840332, "identity": 0.6165575385093689, "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/casual_not_formal/3a0c44ec7489ca3539a97b2faae1d44724363ce33950852dd59f066ccf7eba0e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "3a25c2f8ffcf8a255d458215fb69c6f97c352e59f9c1ec8fc35e1a509b8db8aa", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d1f2eec35daf3573ee8fab0066f38e6da2a0f75e786f336cc621a3f3aa9f78a8", "generation_seconds_batch": 17.309360736981034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9064013957977295, "target_z": -1.5195312535984487, "ce": 6.308549880981445, "genuineness": 4.419482707977295, "identity": 0.5337157845497131, "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/casual_not_formal/3a25c2f8ffcf8a255d458215fb69c6f97c352e59f9c1ec8fc35e1a509b8db8aa.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": "3b019e51b51ebc2b6134f7edeab804b225fc1c5b8ba4b1729ff70336fb36d405", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "ea03542dff20cb7ae42fcb30c20b2d74b1931fd51d4e03bf84b3b11275a30abe", "generation_seconds_batch": 17.74569149094168, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2495837211608887, "target_z": -0.5869139856375633, "ce": 5.824047088623047, "genuineness": 4.457139015197754, "identity": 0.5816516280174255, "hypothesis": "J'ai écrit le numéro sur l'enveloppe. L'e-le bien et vérifie les deux derniers chiffres. Appelle-moi quand tu auras un moment.", "wer": 0.045454545454545456, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/3b019e51b51ebc2b6134f7edeab804b225fc1c5b8ba4b1729ff70336fb36d405.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "3b3da103bd8cfeb25282570a58b326b9cee6e1036bddfee6482b5412701eb1be", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bd959ab40b59ff71d413af2dada68b1a8078be3ce587ea7c8ddda0bfc7ea812", "generation_seconds_batch": 21.46840405999683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2983558177947998, "target_z": -2.0859375120503256, "ce": 6.443574905395508, "genuineness": 4.13078498840332, "identity": 0.6206802129745483, "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/casual_not_formal/3b3da103bd8cfeb25282570a58b326b9cee6e1036bddfee6482b5412701eb1be.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "3b3e35605c16d744b363ba2bf89849064168752bb46363951b844b13b54fcd45", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8c4f9a753c5933638b2ab19934ac226d94255d8caa943c78b63d1c493e155859", "generation_seconds_batch": 16.862784418975934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9630048274993896, "target_z": -0.5488281174245275, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.3768562078475952, "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/casual_not_formal/3b3e35605c16d744b363ba2bf89849064168752bb46363951b844b13b54fcd45.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "3b90b18cffd5fe2e6c527ad8bb41f8d4b8011e3b2e9ef18aadb713842c2da75c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de1837e8d17a27ef20a0d02360651495a068b2572ecddbdf6fd7b49003feead2", "generation_seconds_batch": 16.597163792001083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.856561541557312, "target_z": -1.546874994413439, "ce": 5.816103935241699, "genuineness": 4.13078498840332, "identity": 0.7408517599105835, "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/casual_not_formal/3b90b18cffd5fe2e6c527ad8bb41f8d4b8011e3b2e9ef18aadb713842c2da75c.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": "3bc3b7cc1b6c21034b890edbe33b86201426eb157fd6568a01f56828b3381102", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 128, "seconds_generated": 10.24, "mp3_sha256": "95d6f9f34d89945b2deb3f4eccb4f03ef9d0b77791cd0cf82e3ffe5f9dd66ca4", "generation_seconds_batch": 15.970739534008317, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6839020252227783, "target_z": -1.0898437495902122, "ce": 6.09409761428833, "genuineness": 4.319066047668457, "identity": 0.7666146755218506, "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/casual_not_formal/3bc3b7cc1b6c21034b890edbe33b86201426eb157fd6568a01f56828b3381102.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": "3c241d50b42c280333471078baadb8a1faef1933bf31036bd7d90fe546f6c0db", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51c4dcb2ec80ce7ae0633e1c99dd5f1ab4586c50c27b632a2af5f34cf14a4d90", "generation_seconds_batch": 23.89727048401255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5698041915893555, "target_z": -0.2075195969705288, "ce": 6.157639026641846, "genuineness": 4.337893962860107, "identity": 0.7318386435508728, "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/casual_not_formal/3c241d50b42c280333471078baadb8a1faef1933bf31036bd7d90fe546f6c0db.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "3c579f3889ece8ebf31d8475ba7abf44c6d7f82d3b7e4c0717da8313ecdbf45f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "599e9489dfd466441bab4001d920e55c349f2d0c31f0042b19221bf87f8086bd", "generation_seconds_batch": 23.542789327912033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.987924814224243, "target_z": -0.6503906129752866, "ce": 6.316492557525635, "genuineness": 4.111956596374512, "identity": 0.42502474784851074, "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/casual_not_formal/3c579f3889ece8ebf31d8475ba7abf44c6d7f82d3b7e4c0717da8313ecdbf45f.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "0556855cef4ad6b54e7d763e7b217b2511f45ff9c43ca313e0b955bc8fc43fb1", "generation_seconds_batch": 15.911555426078849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3991029262542725, "target_z": -0.45605470214111277, "ce": 6.340320587158203, "genuineness": 4.432034492492676, "identity": 0.6958703398704529, "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/casual_not_formal/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "3c5a357a6460bf23c2d238b403454afe9ca5234877b9bbdd41de1306b8b5bf7e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d32d841f03577563a11ae65aa8209b229a25d61cd417d5fd53237073fc8f407", "generation_seconds_batch": 17.373828989919275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.295151710510254, "target_z": -1.3359374203154706, "ce": 6.427689552307129, "genuineness": 4.331617832183838, "identity": 0.6067564487457275, "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/casual_not_formal/3c5a357a6460bf23c2d238b403454afe9ca5234877b9bbdd41de1306b8b5bf7e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "3c65675c66adc850e5bc8d7f6f5552575e11301c69e2f43450059f24b453c386", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03e421ec1b02a90fb284e196ef8001eb6a9e9cee55b6ce744177208079494e30", "generation_seconds_batch": 20.2761810059892, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7088217735290527, "target_z": -0.8027344954376134, "ce": 6.149696350097656, "genuineness": 4.193545341491699, "identity": 0.7072242498397827, "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/casual_not_formal/3c65675c66adc850e5bc8d7f6f5552575e11301c69e2f43450059f24b453c386.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "3c75630edbc182ae034c71b34668bbead220eddb7638ba9ab0837c9d93b584c9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "8fd416b6378b4ff14841c10f33f498a6c14b667fff76388ecdc2db620d36894b", "generation_seconds_batch": 18.651032480993308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6763367652893066, "target_z": -0.045166006844775716, "ce": 5.9431867599487305, "genuineness": 4.413206577301025, "identity": 0.7595472931861877, "hypothesis": "Die Tür ist noch offen. Ich hab deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "wer": 0.05263157894736842, "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/casual_not_formal/3c75630edbc182ae034c71b34668bbead220eddb7638ba9ab0837c9d93b584c9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "3cb0e37cc80b218223868e6d955d1d8dce8b6dc5b06df90296b12a23cdf048c0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e282bf2d6bd8abebda6abc2565a07703ab9ded4b10497cdf407105d1bc90f6f", "generation_seconds_batch": 24.746613061986864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.348550796508789, "target_z": 0.30175777212348814, "ce": 6.157639026641846, "genuineness": 4.406930446624756, "identity": 0.6682015657424927, "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/casual_not_formal/3cb0e37cc80b218223868e6d955d1d8dce8b6dc5b06df90296b12a23cdf048c0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "3ce1b37efb927072446bdf776bdfbfe67bffaa51cc8add2997aefdd0ee248404", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f001dde19c4d4d39e52cb270555825490181c53cf71ca58271112fe270535695", "generation_seconds_batch": 25.347691197064705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.398390769958496, "target_z": 0.2995605659076253, "ce": 6.070269584655762, "genuineness": 4.519899368286133, "identity": 0.7847856879234314, "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/casual_not_formal/3ce1b37efb927072446bdf776bdfbfe67bffaa51cc8add2997aefdd0ee248404.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": "3cf35308c065c7350ed10469bb7f61ffd5ba5603b65bc5aafbc177ac5a58b1fd", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a134344e935f131e4ad27a79a6e2d1808488977830dc93eda7b8478557716a5", "generation_seconds_batch": 16.770892351982184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3168673515319824, "target_z": -0.5820312867104287, "ce": 6.149696350097656, "genuineness": 4.425758361816406, "identity": 0.7630963325500488, "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/casual_not_formal/3cf35308c065c7350ed10469bb7f61ffd5ba5603b65bc5aafbc177ac5a58b1fd.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "3d2c7aebefbe58e9a137ee94141e2ef3e1c0bc065fa10c7faaaf4c99cfa45265", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55760b98ebc399b2f6980d2f8825fda32566eb1e4a0280a88cd2ceed0a52f8ae", "generation_seconds_batch": 23.34196627791971, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.6484375173023326, "ce": 6.197352409362793, "genuineness": 4.218649387359619, "identity": 0.3987184464931488, "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/casual_not_formal/3d2c7aebefbe58e9a137ee94141e2ef3e1c0bc065fa10c7faaaf4c99cfa45265.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "3d6bb5f04df330edb0a7dbd95ebd78c02af1197f66e0e8ebe1d1ab7b485e7134", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f23f0b1191a93c756e27b7fe86eab304cf6f3daaef35921f6d6e873418691284", "generation_seconds_batch": 23.158721837913617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.554318428039551, "target_z": -0.8906251167319148, "ce": 6.109982967376709, "genuineness": 4.2374773025512695, "identity": 0.8190388083457947, "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/casual_not_formal/3d6bb5f04df330edb0a7dbd95ebd78c02af1197f66e0e8ebe1d1ab7b485e7134.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "3d8baeac23bb33d37487d376fd4ebe1a7269e13fa822caf41c3d62f203895766", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dfb973c0d3c774116fdf8031c61c7c1b944360dd6d0d3477a8a5c291426422ea", "generation_seconds_batch": 19.9915918110637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9660305976867676, "target_z": 0.07958984192396312, "ce": 5.5738525390625, "genuineness": 4.262581825256348, "identity": 0.6160936951637268, "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/casual_not_formal/3d8baeac23bb33d37487d376fd4ebe1a7269e13fa822caf41c3d62f203895766.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "3e22c3c54b71ca1b97c5d04c26d4f25e90f8cd8a0e648a6df6c2318c1f6a5c10", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d59c2688955b0a6a9b86180ab8fe563a8af599fcc59306f3b362bc4828478af", "generation_seconds_batch": 17.648491839063354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9961129426956177, "target_z": -1.4765625667986222, "ce": 5.935244083404541, "genuineness": 4.143336772918701, "identity": 0.3860004246234894, "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/casual_not_formal/3e22c3c54b71ca1b97c5d04c26d4f25e90f8cd8a0e648a6df6c2318c1f6a5c10.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "3e74389162e42034e1449bc90d5648c51b9497b0e5ade85634877876cb3d1968", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a36e8a4be7f622c3d79307f7c81f095a7a064ad3f9c7becbf4cc9a8af1542858", "generation_seconds_batch": 18.07277487590909, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8964333534240723, "target_z": -1.5976563080538135, "ce": 6.1020402908325195, "genuineness": 4.193545341491699, "identity": 0.4536608159542084, "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/casual_not_formal/3e74389162e42034e1449bc90d5648c51b9497b0e5ade85634877876cb3d1968.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "3e82f16e2b7682282652dc8c45dc54fae1c02ee417579d7341daace1fa63c336", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9fc56a4fd30d9a20badc43c95ee0a844294aa7544d3d5c1905855af1d07d051f", "generation_seconds_batch": 19.615298950928263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0975723266601562, "target_z": -0.6757812348242423, "ce": 6.205295085906982, "genuineness": 4.6705241203308105, "identity": 0.49825212359428406, "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/casual_not_formal/3e82f16e2b7682282652dc8c45dc54fae1c02ee417579d7341daace1fa63c336.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "3eceda4e7f19243430437f4f4f1db7f8377619c7c25dfd7da2d73010c6f4bcf7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ffcdccb1ce9f528a030da6bad71e50e69ddb36354a692d8347e104afb7f4c0b", "generation_seconds_batch": 22.021132701891474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5941903591156006, "target_z": -1.085937575468197, "ce": 5.9431867599487305, "genuineness": 4.15588903427124, "identity": 0.6375898122787476, "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/casual_not_formal/3eceda4e7f19243430437f4f4f1db7f8377619c7c25dfd7da2d73010c6f4bcf7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "3f9a1200e5f4a248b8d0631305f2faa74010540246a0e23a89f86746731837b6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4b04286134ac4fcde541cfde3e84b44abd283a4201f48d6c6ddaa5389b4c381c", "generation_seconds_batch": 17.678491902071983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2720115184783936, "target_z": -0.3320312567631005, "ce": 5.577824115753174, "genuineness": 4.13078498840332, "identity": 0.5874700546264648, "hypothesis": "Heb je Jas op de stoel gelegd? Zeg maar wanneer je klaar bent.", "wer": 0.3157894736842105, "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/casual_not_formal/3f9a1200e5f4a248b8d0631305f2faa74010540246a0e23a89f86746731837b6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "3fff6e02298e5709a31be0029bdbdeb7886d860d5706903a4b5012d98ea88474", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "6b12375fc56f1c68b4acc6022741d2ed362821010a7e96b81d3eabb5bc3438b7", "generation_seconds_batch": 16.529483002028428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -1.2382811808995229, "ce": 6.149696350097656, "genuineness": 4.300237655639648, "identity": 0.7910677790641785, "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": 0.120000000000001, "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/casual_not_formal/3fff6e02298e5709a31be0029bdbdeb7886d860d5706903a4b5012d98ea88474.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "40284eb88e4001521195275d4d397ebdb85e291cca3246afdfd7345884ac27a5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0453f3564b2b692cf5526d9292d5b38df0bce0d89b6912c55735ed31787704ea", "generation_seconds_batch": 20.963755113072693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6390461921691895, "target_z": -0.8535155923338256, "ce": 6.213237762451172, "genuineness": 4.149612903594971, "identity": 0.5946276187896729, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschreien.", "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/casual_not_formal/40284eb88e4001521195275d4d397ebdb85e291cca3246afdfd7345884ac27a5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "4038a76af8ae09b81e45c00a7c416601c735c31d2898391e14926ea1773967d1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "964e2110d950c2006235f5c28ec6ed9ff856c8b578da69ad9cf7d4998d5e4f5b", "generation_seconds_batch": 26.556784225977026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7369461059570312, "target_z": -1.7460937474532319, "ce": 6.38797664642334, "genuineness": 4.3504462242126465, "identity": 0.5583854913711548, "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/casual_not_formal/4038a76af8ae09b81e45c00a7c416601c735c31d2898391e14926ea1773967d1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "4057a427f38d756e5b3a037202205981fd3ceae32275c03532acc669d2dbd1d9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cbe01414425199f6b5e576eaa6491a9c612bbd000e42b0b71e9d19f62da60924", "generation_seconds_batch": 19.615298950928263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0160489082336426, "target_z": -1.5898437259713842, "ce": 6.1020402908325195, "genuineness": 4.118232727050781, "identity": 0.7468183040618896, "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/casual_not_formal/4057a427f38d756e5b3a037202205981fd3ceae32275c03532acc669d2dbd1d9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "406635f5d10c11bdcd4d3495430de98e6968c2d01b7b3b1e59f1ef7abe0d8f79", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "de4ba9e584ab4ae84b5ded5791803406ea49837839bc9bf51f748843e403f9e9", "generation_seconds_batch": 24.680514305015095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0128445625305176, "target_z": -0.789062469355553, "ce": 6.24500846862793, "genuineness": 4.331617832183838, "identity": 0.6152803301811218, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/406635f5d10c11bdcd4d3495430de98e6968c2d01b7b3b1e59f1ef7abe0d8f79.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "4072cedc31dae3f21e92f414a219601a35345156668ad5c86d51195ad1a86ba0", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18fd66a8a58fcb533255815c7865d87f7d281989bee573dc2ec8b9f191746bd3", "generation_seconds_batch": 15.40404078294523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0527164936065674, "target_z": -0.35839842909094466, "ce": 5.959072113037109, "genuineness": 4.231201171875, "identity": 0.6863229274749756, "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/casual_not_formal/4072cedc31dae3f21e92f414a219601a35345156668ad5c86d51195ad1a86ba0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "407bb18445d788a1cfeea5e99b8eba77d36aded8cb4f4ca52e854fb94f5a2f52", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adf7e10b15ab4c02df198bb53ba535fece7030eabf800499d5f0e3b8b9f8154b", "generation_seconds_batch": 16.702504288055934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7355215549468994, "target_z": -0.16992190950541813, "ce": 6.348263263702393, "genuineness": 4.3504462242126465, "identity": 0.4614622890949249, "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/casual_not_formal/407bb18445d788a1cfeea5e99b8eba77d36aded8cb4f4ca52e854fb94f5a2f52.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "40e52ef3dd06a3171f9c182abce73c61e3cd16b16eec8ced6431725189872a23", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0b94375ebc0d40cd03b89f5619ad2f11c7997eb22023b4cbb489cc1ba0e1ce5", "generation_seconds_batch": 23.193395684007555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9161908626556396, "target_z": -0.0576171570930598, "ce": 5.863759994506836, "genuineness": 4.394378662109375, "identity": 0.6285799741744995, "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/casual_not_formal/40e52ef3dd06a3171f9c182abce73c61e3cd16b16eec8ced6431725189872a23.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": "4150cc1fe6d05df24ed4cfcb2aeefd602b5c5fa884e7b35abf636459dce905b2", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be59159a072f37fcce2a12e29ce6f545afd02fbc6e880c6000bb21be973fdd23", "generation_seconds_batch": 22.750562725006603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3492631912231445, "target_z": -0.5400390623415275, "ce": 6.356205940246582, "genuineness": 4.18099308013916, "identity": 0.2506592273712158, "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/casual_not_formal/4150cc1fe6d05df24ed4cfcb2aeefd602b5c5fa884e7b35abf636459dce905b2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "416a94cd36269c2cce51643a91573710a5dc922d9181c99dff99619eff57e263", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62f9e7e96bf8a2f35e7cff9e07511dc41927783c30eea270201658a6777a209a", "generation_seconds_batch": 23.185679672984406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.763089179992676, "target_z": -0.22747805334336899, "ce": 5.990842819213867, "genuineness": 4.582659721374512, "identity": 0.6691588163375854, "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/casual_not_formal/416a94cd36269c2cce51643a91573710a5dc922d9181c99dff99619eff57e263.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "4182d75c595f835b28fe0ce7ae01cc378135d999f0dd026c3aeeb5a4dd8eb857", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4ca0ee703e86611d9022568b3d42a9c7c1aba54df279e03d217a77dfa173e7a6", "generation_seconds_batch": 22.042742638965137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.92313289642334, "target_z": -0.5253906549937516, "ce": 6.1894097328186035, "genuineness": 4.20609712600708, "identity": 0.307035356760025, "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/casual_not_formal/4182d75c595f835b28fe0ce7ae01cc378135d999f0dd026c3aeeb5a4dd8eb857.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "41a3403ba5a1d0502b378109b4a057d0239718f78359cbaca33f9b8309def22c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b5703f5faa1c8818230c23f897aa7c8a5862e8a29a337be2e8a68fa07a9818f", "generation_seconds_batch": 26.462575230980292, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.333599090576172, "target_z": 0.3222656451660456, "ce": 6.181467056274414, "genuineness": 4.65797233581543, "identity": 0.5294508934020996, "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/casual_not_formal/41a3403ba5a1d0502b378109b4a057d0239718f78359cbaca33f9b8309def22c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "41b9559dd4a92b1b891c197e2649c874b6a17cb833a35f53dd6463e5e61dd6b7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f5cb934359069c5bb23422bd89d992efea04e6f7d5e8ce875dc64a707162e31b", "generation_seconds_batch": 23.368959016981535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.842677116394043, "target_z": -0.10253902707656103, "ce": 6.181467056274414, "genuineness": 4.362998008728027, "identity": 0.6531630754470825, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/41b9559dd4a92b1b891c197e2649c874b6a17cb833a35f53dd6463e5e61dd6b7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "42003cc22413e5821c5ea7802a46dbba417a2d501b4b462f8c0dda2997108ab5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adeb9ea1c0a407df6248f5918d76d45ef6d0b8b5c36c0d99e773c91d3a1c7af3", "generation_seconds_batch": 25.05234414502047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9380850791931152, "target_z": -0.7636717756712637, "ce": 5.967014789581299, "genuineness": 4.319066047668457, "identity": 0.32921600341796875, "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/casual_not_formal/42003cc22413e5821c5ea7802a46dbba417a2d501b4b462f8c0dda2997108ab5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "42155cc97f077992bc8045b8a24a289da26a69845036a81d84cdeb4100bf8218", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "797974763e7e2bc7ee1e925098bf94e38583f007f6b25b6509c951d32050bc31", "generation_seconds_batch": 26.366761818062514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301915407180786, "target_z": -0.5039062533295263, "ce": 6.086154937744141, "genuineness": 4.557555198669434, "identity": 0.6794254779815674, "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/casual_not_formal/42155cc97f077992bc8045b8a24a289da26a69845036a81d84cdeb4100bf8218.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "42aac5b6299c8579abf33090734ea292ab299311010aece8f7a3059b66cf2e14", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87dc7d12a9a09ccdf15f0fa4022619e23147d258aae7cfed537e3f2121e668ac", "generation_seconds_batch": 26.06323558499571, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195472240447998, "target_z": -1.3281249710715366, "ce": 6.229123115539551, "genuineness": 4.231201171875, "identity": 0.3191595673561096, "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/casual_not_formal/42aac5b6299c8579abf33090734ea292ab299311010aece8f7a3059b66cf2e14.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "4313337c6b04537c838572b1f727ea127307e78f821b599a0bd771abc33b0d50", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d368dac988933b13f0776e52fd3691b5729f9b4d1b6d908d2c31ef9fc6160c90", "generation_seconds_batch": 16.794543407973833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5375869274139404, "target_z": -1.5546875363046424, "ce": 6.133810997009277, "genuineness": 4.099404335021973, "identity": 0.6594083905220032, "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/casual_not_formal/4313337c6b04537c838572b1f727ea127307e78f821b599a0bd771abc33b0d50.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "e6edd1ec229bea6eb3b31e3d24cd88188e8520de6027f27649c8c5298d1149ae", "generation_seconds_batch": 17.21664079092443, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3150875568389893, "target_z": -1.1601562425805994, "ce": 6.054384231567383, "genuineness": 4.143336772918701, "identity": 0.5629101395606995, "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/casual_not_formal/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "43888a6ee08f91467ae5ca26a674d7ba8e36f1d27f026dc049b22d13b78e073c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e212cf92c686f98436bc58713ec8dcee11539e7ddf2519c4ade6ba6802ff2ac3", "generation_seconds_batch": 24.16852970002219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.604158401489258, "target_z": -1.1679687608529747, "ce": 6.38003396987915, "genuineness": 4.325342178344727, "identity": 0.525420069694519, "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/casual_not_formal/43888a6ee08f91467ae5ca26a674d7ba8e36f1d27f026dc049b22d13b78e073c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "43bc36fddc668ba207747a6fba141c399569fb10afef4b9dcfd0292e246bf55c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bbf4a014ee135b0cece3ca1cc5ab03d64d816ac64f8938a3407a4f081a24d77", "generation_seconds_batch": 24.86681206803769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.887532949447632, "target_z": 0.009277364119538313, "ce": 6.109982967376709, "genuineness": 4.331617832183838, "identity": 0.6905765533447266, "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/casual_not_formal/43bc36fddc668ba207747a6fba141c399569fb10afef4b9dcfd0292e246bf55c.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "a77ad87d435a3234a6173e090c1aa71bfde405af06ffaa37ed5c7199224a2c7d", "generation_seconds_batch": 17.25446208298672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0676684379577637, "target_z": -0.7441406571299285, "ce": 6.197352409362793, "genuineness": 4.582659721374512, "identity": 0.6661312580108643, "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/casual_not_formal/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "4434633575046ad4072a114aa6750e899ff50a9cc8d1780aba6ee0dd7d73a462", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e90fcacc559fd25071fbe9b8cf5e69d78427ae7fbff3b1eabbfc4dcf0506188d", "generation_seconds_batch": 18.945967592997476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5311784744262695, "target_z": -0.5332031086733149, "ce": 6.141753673553467, "genuineness": 4.7081804275512695, "identity": 0.7363121509552002, "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/casual_not_formal/4434633575046ad4072a114aa6750e899ff50a9cc8d1780aba6ee0dd7d73a462.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4452204ed2f327573c2a5e4c92067a3d30b5c480c98f9b8ab808e3475460d15e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d67b6dab9b360c06766de31fca766cee8624ce40ea02f969879f3ff31ff69e48", "generation_seconds_batch": 21.32025559002068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.321139335632324, "target_z": 0.21801769879202026, "ce": 6.292664527893066, "genuineness": 4.406930446624756, "identity": 0.45538440346717834, "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/casual_not_formal/4452204ed2f327573c2a5e4c92067a3d30b5c480c98f9b8ab808e3475460d15e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "45aecef61eebb6a1431b7486a1077554f4f1ba0a47a246e53537bcce715cc21c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44842397bc52b063f2c39174796344e1c6c84a4d86da0e5ac7290260f8119aa7", "generation_seconds_batch": 17.588289493927732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.873293161392212, "target_z": -0.9296875539863494, "ce": 6.205295085906982, "genuineness": 4.432034492492676, "identity": 0.6396538019180298, "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/casual_not_formal/45aecef61eebb6a1431b7486a1077554f4f1ba0a47a246e53537bcce715cc21c.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "de43e3aaf4a661086426a662afec0ca02c13002b3c1eb506a942d18e7d0913a9", "generation_seconds_batch": 15.921528177917935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4777791500091553, "target_z": -1.9921875801167275, "ce": 6.332377910614014, "genuineness": 4.118232727050781, "identity": 0.546562910079956, "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/casual_not_formal/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "469ea3384913243888bc764b4279de8aecd813aa53485d481bdf43fdbe8dbb12", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e4db62312ebbf2edc699fac6a68c4c22253c7b1fb144a4378d2cb902b29b85e1", "generation_seconds_batch": 26.50834251602646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.765624927846922, "ce": 6.117925643920898, "genuineness": 4.319066047668457, "identity": 0.30596432089805603, "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/casual_not_formal/469ea3384913243888bc764b4279de8aecd813aa53485d481bdf43fdbe8dbb12.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "46bd4bdd94ac0fc3ab91b1c15cf520398abb4b79f5236997c87d8ef588c89792", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3750280f4859ac020fc11a23ff5f958062dbde6f2335e6655b84b6864aca29b0", "generation_seconds_batch": 16.099584136041813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.444671154022217, "target_z": -1.1640625435383574, "ce": 5.839932441711426, "genuineness": 4.105680465698242, "identity": 0.5822001695632935, "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/casual_not_formal/46bd4bdd94ac0fc3ab91b1c15cf520398abb4b79f5236997c87d8ef588c89792.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": "46d73e45b68db75ab977b4bcf58feb21f8612cb695e17ba64dc3edfaa90458f9", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "8de2654bf10cffd237b176fbda5610e65f6cedcccfb169773cb791c227109d85", "generation_seconds_batch": 15.541306035942398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.124271869659424, "target_z": 0.2714843295679724, "ce": 6.252951145172119, "genuineness": 4.494794845581055, "identity": 0.740420937538147, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.09523809523809523, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/46d73e45b68db75ab977b4bcf58feb21f8612cb695e17ba64dc3edfaa90458f9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "46ddf5f65bc8c12eeb03d942afc6001407c52834459e714db1dc2ee6b056c028", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7a6b68f4707a858252bd0574e852984aa4b40a66f257c807b150b77dd19dbcc9", "generation_seconds_batch": 19.106700982898474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.821495294570923, "target_z": 0.009887706880654563, "ce": 6.173524379730225, "genuineness": 4.413206577301025, "identity": 0.5983350872993469, "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/casual_not_formal/46ddf5f65bc8c12eeb03d942afc6001407c52834459e714db1dc2ee6b056c028.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "46e88e2134148a3123131e339a1f71dbb3830b25e5397c1c80577ba300e0f317", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02d532c9908d890174ebc8e2f5fdf30b0a31dd027b838c2a41e3902250cb234b", "generation_seconds_batch": 16.641121591092087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291947364807129, "target_z": -0.43554694063569, "ce": 5.816103935241699, "genuineness": 4.557555198669434, "identity": 0.8423103094100952, "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/casual_not_formal/46e88e2134148a3123131e339a1f71dbb3830b25e5397c1c80577ba300e0f317.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "471460b6bc1154889d1d4c40c2ae995e07bba40977b4cf785fcc26bf83416b8d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bcf15d1bac588b568eed93af488c631828befbc967d0ad0b0df5096fbc5965d5", "generation_seconds_batch": 18.134781057946384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.2031249253353722, "ce": 6.141753673553467, "genuineness": 4.300237655639648, "identity": 0.5186789035797119, "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/casual_not_formal/471460b6bc1154889d1d4c40c2ae995e07bba40977b4cf785fcc26bf83416b8d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "47722f711f93f96533517b1780dd5ee7b1858aede8a8977977c5dfc01e775b1e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f463fbf551cda73a0b871003f489d6e2e0fc55b12bfebe1cfa0912a7f5927a3c", "generation_seconds_batch": 15.738997455919161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3118834495544434, "target_z": -0.5781249738119936, "ce": 6.141753673553467, "genuineness": 4.6705241203308105, "identity": 0.7210925817489624, "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/casual_not_formal/47722f711f93f96533517b1780dd5ee7b1858aede8a8977977c5dfc01e775b1e.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "78121af245af089ccc884eee9d386a7867785047b1585d438ea860ade83f14fe", "generation_seconds_batch": 16.048109869938344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -1.1093748756998165, "ce": 6.133810997009277, "genuineness": 4.6705241203308105, "identity": 0.7303548455238342, "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/casual_not_formal/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "47e7904cb60adbecb8ee6cb3db414d644371d115fae5a5f17f864360009f731e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2b851ba2ed7b26175fa571d76ef30097a920973bc9891d63d94dabc32fd888d1", "generation_seconds_batch": 24.16852970002219, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3250555992126465, "target_z": -1.34765629526555, "ce": 6.459460258483887, "genuineness": 4.356721878051758, "identity": 0.5347812175750732, "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/casual_not_formal/47e7904cb60adbecb8ee6cb3db414d644371d115fae5a5f17f864360009f731e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "48682a79c0c262b397f932ad63cd783aad5f2487ecc4fcee8b857e47a7bace42", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d75962ee0de0c27b964e5fd460c7381e7de0bc48c123b52a58b457013571a077", "generation_seconds_batch": 23.022883794968948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.702191114425659, "target_z": -0.048461927506164224, "ce": 5.974957466125488, "genuineness": 4.268857479095459, "identity": 0.6620264053344727, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/48682a79c0c262b397f932ad63cd783aad5f2487ecc4fcee8b857e47a7bace42.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "489de33c84313230d53263c2c169f5d504c73a187cda9488a74767a5f343db08", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "35339ee4e7bb925fe3d6f779acb2bf3667cc1590984c58fdf56caf2cd45c00a3", "generation_seconds_batch": 24.447202009963803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269519567489624, "target_z": -0.47753905513182193, "ce": 6.1020402908325195, "genuineness": 4.720732688903809, "identity": 0.7812894582748413, "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/casual_not_formal/489de33c84313230d53263c2c169f5d504c73a187cda9488a74767a5f343db08.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "48b140b5dabbcbb0f50d10d2d6d48c245743ee617815dc1a306485b97c6a7de0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a8b5eba48cc48b9d751a8290e51e090102d1cd369b7c440421858519e0c15f5", "generation_seconds_batch": 17.841417910065502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6744678020477295, "target_z": -0.19287110796811818, "ce": 5.804190158843994, "genuineness": 4.030367851257324, "identity": 0.7905246019363403, "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/casual_not_formal/48b140b5dabbcbb0f50d10d2d6d48c245743ee617815dc1a306485b97c6a7de0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4927a509fd32aa0a54c2f4cc1423e2d4a4eb9ab24fc588af04948f35b2cbc887", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b2097f28a6e472d56f7bab7819e9eb2904f51c971ee005622a298ef412c1ca9", "generation_seconds_batch": 26.39916119107511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246379375457764, "target_z": 0.23144531700846652, "ce": 6.38003396987915, "genuineness": 4.557555198669434, "identity": 0.5814538598060608, "hypothesis": "I've found the little box. It was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/4927a509fd32aa0a54c2f4cc1423e2d4a4eb9ab24fc588af04948f35b2cbc887.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4930c873a16a64d5a4cdf8671374e476da5a1b06438247d3af48fa94099818fd", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f4eba0cb86934d2b70b4e538080f8cfb3f64b4483f97eccdd70a06cc8bfb575", "generation_seconds_batch": 16.216135946917348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4165468215942383, "target_z": -0.44140631169148214, "ce": 6.038498878479004, "genuineness": 4.595211505889893, "identity": 0.6304502487182617, "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/casual_not_formal/4930c873a16a64d5a4cdf8671374e476da5a1b06438247d3af48fa94099818fd.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "493c2a4ba08ad7b4ac5ee5f355a1b1e186da74e4b5f0e016b0f5b41df6724020", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f328cdc9b5bde99c24a86d8825fc7ad899082759e7c541fe5ba567da34b0765c", "generation_seconds_batch": 21.599056101986207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547910213470459, "target_z": 0.4960937936472104, "ce": 6.316492557525635, "genuineness": 4.3881025314331055, "identity": 0.40292859077453613, "hypothesis": "I found the little box. It was the side the old window. We can look inside together.", "wer": 0.125, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/493c2a4ba08ad7b4ac5ee5f355a1b1e186da74e4b5f0e016b0f5b41df6724020.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "495d82c678f6c9786069520dc28cc19d091343e8caacaf20b55c0f7063d65889", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3ce42d8c4e8ab22414890ad2dcbe21bfd761f88a59abb16f5911cae0548e923", "generation_seconds_batch": 25.824389254907146, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.713805913925171, "target_z": -0.8359375173999763, "ce": 6.419747352600098, "genuineness": 4.532451152801514, "identity": 0.6225656270980835, "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/casual_not_formal/495d82c678f6c9786069520dc28cc19d091343e8caacaf20b55c0f7063d65889.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "4968ac5c549c1611904c188b1d09bf466a81bf73fe691e41f123ea5687e591f5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c78c2e99ea3ec63fffca5ab01a3a1fbd17791b68d7ad6631217538725f0eed29", "generation_seconds_batch": 23.913284947979264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.356739044189453, "target_z": -0.34375002699557733, "ce": 5.974957466125488, "genuineness": 4.118232727050781, "identity": 0.6371268033981323, "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/casual_not_formal/4968ac5c549c1611904c188b1d09bf466a81bf73fe691e41f123ea5687e591f5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "49a983ac9b305fc4330a4bb17aa96772c7cf0feefeb7e1e2808e149bea2b6efc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d55b978ea417a4372f50decd196ad1538187b12e4f4779d74e32311d5f33c7a1", "generation_seconds_batch": 22.578925330075435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.29871129989624, "target_z": 0.2308349885857265, "ce": 5.95112943649292, "genuineness": 4.482243061065674, "identity": 0.49291568994522095, "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/casual_not_formal/49a983ac9b305fc4330a4bb17aa96772c7cf0feefeb7e1e2808e149bea2b6efc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "49c465e1e483655c1741a47bc4ac8aa4497f8d61cdc9e68f55740f528b01be4b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d321ca5bc6cfb673ad0584ef702567562c8e7771d98e14ad6d9cb9d5743c265e", "generation_seconds_batch": 18.04512499505654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5575227737426758, "target_z": -1.8164062846798987, "ce": 6.260893821716309, "genuineness": 4.20609712600708, "identity": 0.5901927947998047, "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/casual_not_formal/49c465e1e483655c1741a47bc4ac8aa4497f8d61cdc9e68f55740f528b01be4b.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "5b1a3dfd7287ad1ebecc1c5c2711a5f4610fc2f7ba6a21821d104f8c96295f02", "generation_seconds_batch": 16.16228816693183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.444671154022217, "target_z": -1.1835937057941335, "ce": 6.109982967376709, "genuineness": 4.432034492492676, "identity": 0.7400602698326111, "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/casual_not_formal/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4a2a0749e448826a567b27951bdff227db523466d09f8c81a1034ade55521bb3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1dece559e1e6d5908808808d23805f4732d5c4eb25b8eaf9d7713fa76382c89", "generation_seconds_batch": 26.124393612961285, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1025562286376953, "target_z": -0.6367187543123772, "ce": 6.054384231567383, "genuineness": 4.6203155517578125, "identity": 0.7800103425979614, "hypothesis": "He encontrado una la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/4a2a0749e448826a567b27951bdff227db523466d09f8c81a1034ade55521bb3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4a5c538cde13200fc8ccf43105345c549a29d41ed2ca413cfa2635064ad74961", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "320bdfe2bb2f442505c079b68e5d26391329bf4bf9950546fc0004b036cae5ba", "generation_seconds_batch": 21.827250136062503, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.37841792282714654, "ce": 6.205295085906982, "genuineness": 4.419482707977295, "identity": 0.3019251823425293, "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/casual_not_formal/4a5c538cde13200fc8ccf43105345c549a29d41ed2ca413cfa2635064ad74961.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4ab82dccbdcd801f56f07bc6b8ae2d6f0127769591c4f24d1cc2de9a0d267b59", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0926b8bc4916a12f609a50107256427b9a6d03e078c2c9bba01c54b8a88de4e", "generation_seconds_batch": 22.59894140996039, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.213983535766602, "target_z": 0.3505859201484458, "ce": 5.982900142669678, "genuineness": 4.143336772918701, "identity": 0.657819926738739, "hypothesis": "Ha encontrado la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/4ab82dccbdcd801f56f07bc6b8ae2d6f0127769591c4f24d1cc2de9a0d267b59.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": "4b0b619c5f44c7e291744f27fcc0030f790b88da08151534761a485fbd0c2e63", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d2aa4c16171a2accc2e27bca84c1e925eb4ce19a07fb0336d6e2b9ffec5f1c6", "generation_seconds_batch": 17.25393615302164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.104335784912109, "target_z": 0.07128895780200106, "ce": 6.173524379730225, "genuineness": 4.68307638168335, "identity": 0.6900506615638733, "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/casual_not_formal/4b0b619c5f44c7e291744f27fcc0030f790b88da08151534761a485fbd0c2e63.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "4b0ebf2405e45097aec2e901256ad9af279ab424953cd143b5ec41710fc244ea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bc5432ae5382c580d67a8ac57608fdf9b39f8e55726341e95d17df909d182c5d", "generation_seconds_batch": 24.070664930040948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9761770963668823, "target_z": -1.4921875046933506, "ce": 6.316492557525635, "genuineness": 4.494794845581055, "identity": 0.6189171075820923, "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/casual_not_formal/4b0ebf2405e45097aec2e901256ad9af279ab424953cd143b5ec41710fc244ea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "4b3562c8e2def0d0f486785e213f9e5b98a567b3f5230bed77b5837542d0c378", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae51a1d351acc59c4b85687c98cec6606d5767192c912f032e911f7695213148", "generation_seconds_batch": 22.362677935045213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.713805913925171, "target_z": -0.8437499823022869, "ce": 6.062326908111572, "genuineness": 4.356721878051758, "identity": 0.775587797164917, "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/casual_not_formal/4b3562c8e2def0d0f486785e213f9e5b98a567b3f5230bed77b5837542d0c378.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "4bc9943aca3c1c850d35c21f480939a31e261dccba6bf22781c720da1683e79d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "ef004baacbe2363a3416ed9e7d7e702ddfd70467c0eba9b7d0d6fa51e8a5ee27", "generation_seconds_batch": 24.21990254498087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.500028610229492, "target_z": -0.08935546056744824, "ce": 6.038498878479004, "genuineness": 4.086852550506592, "identity": 0.2877161502838135, "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": 0.120000000000001, "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/casual_not_formal/4bc9943aca3c1c850d35c21f480939a31e261dccba6bf22781c720da1683e79d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4bc99accad423a275ad3525452f342f65075d4fa63f41f5623ce9ccc40d28109", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "59502fb42252979c0c9d9fd00c6d77d53fc862bcf73315b3781d4635771785d7", "generation_seconds_batch": 22.55454672104679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2994234561920166, "target_z": -0.5595702915383902, "ce": 6.221180438995361, "genuineness": 4.218649387359619, "identity": 0.4394643306732178, "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/casual_not_formal/4bc99accad423a275ad3525452f342f65075d4fa63f41f5623ce9ccc40d28109.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "4bda1e0756414f848d9c6d7b39e1b324d203b6d379ed9368063878bb387fef13", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de0edab2194f6b9e8fe191a00e2d4fb456cb1b194277c5257cc9eb7333fc7a77", "generation_seconds_batch": 23.090719264000654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.0976562955264693, "ce": 5.847874641418457, "genuineness": 4.1998209953308105, "identity": 0.5306229591369629, "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/casual_not_formal/4bda1e0756414f848d9c6d7b39e1b324d203b6d379ed9368063878bb387fef13.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4bfce8a4ff57004ce09deff182a356a7c47dcb4376b2d2304cd87b18847c188e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e45b1c84837cd1315369967f7f8dec906aee26f0090db3134e316ba320f3f20b", "generation_seconds_batch": 22.435147687094286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.468878746032715, "target_z": -0.3916015208985283, "ce": 6.181467056274414, "genuineness": 4.256305694580078, "identity": 0.3610919415950775, "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/casual_not_formal/4bfce8a4ff57004ce09deff182a356a7c47dcb4376b2d2304cd87b18847c188e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "4c0739b87c9a0db10e169853ba754b61f8f999ee8a7ec307d76e59f22ec4c9d2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bda70113346b0c8a37a269b42dbaf9593b5837ae93812930fd29a630c0cfe166", "generation_seconds_batch": 22.075880830991082, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.461402654647827, "target_z": 0.003906226709321947, "ce": 5.911416053771973, "genuineness": 4.005263805389404, "identity": 0.3809729218482971, "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/casual_not_formal/4c0739b87c9a0db10e169853ba754b61f8f999ee8a7ec307d76e59f22ec4c9d2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "4c27d3ee14a1c957979dc4fa25db44cfddb08eea4281438a1f91ff250dcc8336", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d2becd9c962c3d537d1b51420c30531b4f6e950356c41402781fa2f9f8102d6", "generation_seconds_batch": 22.68779147497844, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0060808658599854, "target_z": -1.4179688395386472, "ce": 6.0305562019348145, "genuineness": 4.118232727050781, "identity": 0.6196204423904419, "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/casual_not_formal/4c27d3ee14a1c957979dc4fa25db44cfddb08eea4281438a1f91ff250dcc8336.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4c41b593d6ad9c26e4d6924e6c57a7de4bfc409e41b3cf9d20f2c02d1516a245", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e2080783856f921305fef3222d7469eda9183e2a111712fa332bef48649a205", "generation_seconds_batch": 19.10751733998768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.919928789138794, "target_z": -0.04248046323447141, "ce": 5.990842819213867, "genuineness": 4.2751336097717285, "identity": 0.6030492782592773, "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/casual_not_formal/4c41b593d6ad9c26e4d6924e6c57a7de4bfc409e41b3cf9d20f2c02d1516a245.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "4d1c83bcad34e9c7ba491c76d2a683bd96097d7a51b829248f078c754190d15b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bfbe77f88a42f52957ea488ad49a88c9a42302a74758454d3b6694cb44bee8ee", "generation_seconds_batch": 23.30655007099267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.858875036239624, "target_z": -0.18896483930596278, "ce": 6.23706579208374, "genuineness": 4.607763767242432, "identity": 0.7634624242782593, "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/casual_not_formal/4d1c83bcad34e9c7ba491c76d2a683bd96097d7a51b829248f078c754190d15b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "4d5632222a766fcff9cf0435c82346e2a3518a9da6cdb45ffd4f66944acbaa4c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "debca805d2466973230c7c658d1a16bceef18e774576f317cfb983c735981af2", "generation_seconds_batch": 24.74133886594791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.414767265319824, "target_z": -0.9765624989435144, "ce": 6.173524379730225, "genuineness": 4.017816066741943, "identity": 0.31848660111427307, "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/casual_not_formal/4d5632222a766fcff9cf0435c82346e2a3518a9da6cdb45ffd4f66944acbaa4c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4d6995f11b83525bdbdf30da91c420906328f8be72011ff18a8979ed65e66ec2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "016e1e1d195b3e1142d42cfc665c232ce2eec6aea004a0e62b25c3a46c5ac471", "generation_seconds_batch": 23.819262548931874, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6277430057525635, "target_z": -0.17993161591367204, "ce": 6.086154937744141, "genuineness": 4.381826400756836, "identity": 0.6151974201202393, "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/casual_not_formal/4d6995f11b83525bdbdf30da91c420906328f8be72011ff18a8979ed65e66ec2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4d7aec1da4363b0c8d48064b20d30e1c55ccdb523fa2d60b9cadeca8071563b2", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57e2d696bae37ad1ed0870e402277bbea6a6766c379593859b07d376642d5246", "generation_seconds_batch": 18.66885324800387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.799067258834839, "target_z": -0.12158206299815454, "ce": 6.284721851348877, "genuineness": 4.482243061065674, "identity": 0.2608529031276703, "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/casual_not_formal/4d7aec1da4363b0c8d48064b20d30e1c55ccdb523fa2d60b9cadeca8071563b2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "4dd2eb603ce643f725e43ea2ffee234601f4548d4cd9faa16fc865671f7910ee", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b234985ba1380fafe7268fac766144278e9b1d65a36ca38edca01c4a4bc59bfc", "generation_seconds_batch": 25.412665425101295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.670106887817383, "target_z": -0.2443847366409124, "ce": 6.197352409362793, "genuineness": 4.6203155517578125, "identity": 0.6973475217819214, "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/casual_not_formal/4dd2eb603ce643f725e43ea2ffee234601f4548d4cd9faa16fc865671f7910ee.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "4dd44481cd0e2d078c985921d31c5a0a9887bf2a8f277ad41770c192a2c58366", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcf912127e7e45280baa64c5c4d6ed243d89ab8c2c6a1f38eeda3dd3f5be0e70", "generation_seconds_batch": 23.79364049504511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.144531304685871, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.4449412524700165, "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/casual_not_formal/4dd44481cd0e2d078c985921d31c5a0a9887bf2a8f277ad41770c192a2c58366.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "4df22e6ad17ebdb048ae253c2c537ff61c7ce88b2407d7d81e6f26cb50f2c81b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c32e658cf3302e0d6f7fe70d9a3f57c55981151d4cde1ab95d74695bf23ee848", "generation_seconds_batch": 27.056125040980987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3218512535095215, "target_z": -0.4003906715005178, "ce": 6.268836498260498, "genuineness": 4.086852550506592, "identity": 0.6070549488067627, "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/casual_not_formal/4df22e6ad17ebdb048ae253c2c537ff61c7ce88b2407d7d81e6f26cb50f2c81b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "4e3557042160605877d037c7dbf50b2b450734dcb58b33347353fd60aa8265fc", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bef8ab8425e26bc97744b269346b375f4ea37fc3b5cf5de211fee1eae27e8658", "generation_seconds_batch": 17.72676364099607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.145632266998291, "target_z": -1.3554687586023442, "ce": 6.213237762451172, "genuineness": 4.193545341491699, "identity": 0.7916860580444336, "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/casual_not_formal/4e3557042160605877d037c7dbf50b2b450734dcb58b33347353fd60aa8265fc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "4ed92460599542225ee7e1a7d8e5de1698e0cba632bd0cc6797364934a4aa33e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e46099af4fc146417e5c503dd738f6364215f7a527813f7161efcedeb07991a", "generation_seconds_batch": 18.895433694007806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174111843109131, "target_z": 0.27392581133279276, "ce": 6.022613525390625, "genuineness": 4.118232727050781, "identity": 0.3406817317008972, "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/casual_not_formal/4ed92460599542225ee7e1a7d8e5de1698e0cba632bd0cc6797364934a4aa33e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4f30b286b6e5aaa909bc06b5e633f80a962bd6531917780292a780ce78ec06c7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8a0004c0642efdb7e3b7b025bc262ff38215d9c29fc3d30645bd2a497dbbea4", "generation_seconds_batch": 25.887744396924973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.583510160446167, "target_z": -0.3681640811077309, "ce": 6.086154937744141, "genuineness": 4.532451152801514, "identity": 0.6466628313064575, "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/casual_not_formal/4f30b286b6e5aaa909bc06b5e633f80a962bd6531917780292a780ce78ec06c7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4f61d553dfe658ba9c2c1a85629869195c4118913ed5286c13906d439f227a82", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "62716787fe1ce4d7c9b06a7fa351d542e531b851db9984d1ed92ba6dbbed08cc", "generation_seconds_batch": 17.884497439023107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.04274845123291, "target_z": -0.7714844235862738, "ce": 6.205295085906982, "genuineness": 4.457139015197754, "identity": 0.6106749773025513, "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/casual_not_formal/4f61d553dfe658ba9c2c1a85629869195c4118913ed5286c13906d439f227a82.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": "4f81ef7c47c726a6a06db9786fe68ad0d949a54902ca50f057f28d6b2791e591", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91072a56eb9e2208f9e29447170cf8246c26ece268ef341213f512327777f30a", "generation_seconds_batch": 24.068461124086753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.219679832458496, "target_z": -0.6650390481830356, "ce": 6.133810997009277, "genuineness": 4.231201171875, "identity": 0.7959884405136108, "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/casual_not_formal/4f81ef7c47c726a6a06db9786fe68ad0d949a54902ca50f057f28d6b2791e591.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "4fbbd737523c09a43c06daccc8d92cb204af38053dd37493dcffa6d5322b92af", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06cfa0543a8eeffb79499a02e1e445a453aea7ca84570f909ce8234b924bfb00", "generation_seconds_batch": 22.549734112923034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6694839000701904, "target_z": -0.06884764474552524, "ce": 6.213237762451172, "genuineness": 4.231201171875, "identity": 0.2565971314907074, "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/casual_not_formal/4fbbd737523c09a43c06daccc8d92cb204af38053dd37493dcffa6d5322b92af.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "4fce867ad122cd1b653bd6d7c6dac54cde18d89d9207f4300d78511ff24c04d1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec1b630549cf01fb13eff1440b9149fdb435d79f07ff86348d79c20690278562", "generation_seconds_batch": 22.59894140996039, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.037052154541016, "target_z": 0.052734327072026896, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.6316115856170654, "hypothesis": "He encontrado la cajita. Estaba junto a la ventana vieja. ¿Podemos mirar juntos lo que hay de ahí?", "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/casual_not_formal/4fce867ad122cd1b653bd6d7c6dac54cde18d89d9207f4300d78511ff24c04d1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "50e9d39b19151658db024fbcfff57949285e3164722f8fe9de6e06abaf0cc807", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dbde516d6e4f50c7659a800c85b1d307980b58b3e2b64802167bb03e59bdf21b", "generation_seconds_batch": 17.578236727043986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2171878814697266, "target_z": -0.4199218280792888, "ce": 6.173524379730225, "genuineness": 4.7834930419921875, "identity": 0.6996853351593018, "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/casual_not_formal/50e9d39b19151658db024fbcfff57949285e3164722f8fe9de6e06abaf0cc807.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "51105f46a3b9e44d9ec604fd84626dbe81d9a693a54ebc9b2b710127b8bc95dc", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e3c5f86e4ff9d401c09d4a454b2f24bb1a700cf7d099fba16bb18fb04b0393f", "generation_seconds_batch": 18.551569062052295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.9921875172206946, "ce": 6.125868320465088, "genuineness": 4.250029563903809, "identity": 0.6663765907287598, "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/casual_not_formal/51105f46a3b9e44d9ec604fd84626dbe81d9a693a54ebc9b2b710127b8bc95dc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "512a3fe7de07d177c34b186cce54ea89e2767707fa15bd42ee9a3d7d50f067a4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2180f2ff31b7587eeca8fbb0c15e685eb5505c85eab26c51a6d836a46592e5f4", "generation_seconds_batch": 25.209178582997993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.778597593307495, "target_z": -0.9082030760511615, "ce": 6.165581703186035, "genuineness": 4.356721878051758, "identity": 0.31058269739151, "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/casual_not_formal/512a3fe7de07d177c34b186cce54ea89e2767707fa15bd42ee9a3d7d50f067a4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "5141be3ed0cae21d1b1cbd593b51d0dcb00636b3b83db3b641d046407a13dbc3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "534dd99490cab254fda172e48d6ea859dc68d3f5a3d535f1e568c157456cd8d2", "generation_seconds_batch": 20.24067892297171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8663511276245117, "target_z": -0.02685545136762972, "ce": 6.070269584655762, "genuineness": 4.3127899169921875, "identity": 0.27558186650276184, "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/casual_not_formal/5141be3ed0cae21d1b1cbd593b51d0dcb00636b3b83db3b641d046407a13dbc3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "51a00cb91262d34b0b560fa601c6a6f9d2643244d8fc1a4e4ba94f39d26082ed", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c1e47b8c2decb5c3b77b3ba6b1dd9d060fbc624d21303e0dfa00ff33edb6564b", "generation_seconds_batch": 21.952278492040932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.914062513526201, "ce": 6.09409761428833, "genuineness": 4.118232727050781, "identity": 0.5748421549797058, "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/casual_not_formal/51a00cb91262d34b0b560fa601c6a6f9d2643244d8fc1a4e4ba94f39d26082ed.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "52417c8e8f38ed7e653d10a70baf2cea7cd4325a584fcec51e0bccb0ba21f89d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "facf0e84234a7dd201dcd4a34bd92eeb9afa1a47e166c3a7c6ef645947b8c1b1", "generation_seconds_batch": 22.865506227943115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9363051652908325, "target_z": -1.4101562599357553, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.5943777561187744, "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/casual_not_formal/52417c8e8f38ed7e653d10a70baf2cea7cd4325a584fcec51e0bccb0ba21f89d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "52895490ab8dab61e80e0d8d872cf255f76eb94b63bcb39935f1c2f945bb9fcf", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c3ff5fb693427541d87d57738b2a6efa8c110f47bc2a65c155ae860257fe89f4", "generation_seconds_batch": 24.896861342946067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972972869873047, "target_z": -0.7675781479425163, "ce": 6.197352409362793, "genuineness": 4.582659721374512, "identity": 0.6372663378715515, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/52895490ab8dab61e80e0d8d872cf255f76eb94b63bcb39935f1c2f945bb9fcf.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": "531b2034f2294178957644d5f9d3cae7520b30fc72d725c8479fd140e9d54093", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d279ff2b6cef7f6228b6cd51bf3f200693382b0cbc38c41a56385813d397f038", "generation_seconds_batch": 24.465385763091035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.094367980957031, "target_z": 0.13110350260753167, "ce": 6.054384231567383, "genuineness": 4.557555198669434, "identity": 0.4097430109977722, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/531b2034f2294178957644d5f9d3cae7520b30fc72d725c8479fd140e9d54093.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "532d0856b3654350f329a8724eae6b34bb35e7afddb05c9491a2ff0a0b09fc07", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6300404ea94dd2489391e0baa9f77036b4bea092f18c30595aaffeb9c525574", "generation_seconds_batch": 16.477766836993396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.504478931427002, "target_z": -0.9863280808541606, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.750436544418335, "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/casual_not_formal/532d0856b3654350f329a8724eae6b34bb35e7afddb05c9491a2ff0a0b09fc07.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "5381c48d22c2b9ba58cb0f91263d984dac2a9f7d469875c02b9b3bdc7e4b15d2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "546e90b353fc88657615ac817913a752cde43f7b8ed4276b3baea6efdc5c2f6d", "generation_seconds_batch": 24.1828153639799, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8015592098236084, "target_z": -0.16162113362588104, "ce": 6.09409761428833, "genuineness": 4.532451152801514, "identity": 0.7892632484436035, "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/casual_not_formal/5381c48d22c2b9ba58cb0f91263d984dac2a9f7d469875c02b9b3bdc7e4b15d2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "53e606fd4ed55424090ff38a8a6442e5be14350b50e8ebd61a6714ab02336697", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2feae750599309bfcff0302adfa286743068bb69f052fd610b6d9e8df39eaa5c", "generation_seconds_batch": 25.33752108097542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0559206008911133, "target_z": -1.5078125918693164, "ce": 6.324435234069824, "genuineness": 4.262581825256348, "identity": 0.6884570121765137, "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/casual_not_formal/53e606fd4ed55424090ff38a8a6442e5be14350b50e8ebd61a6714ab02336697.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": "541757d064ffa7556ce493ab560e49e4cee28d666e7b6a05bc30d0c14335945e", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a57902f9f6d60559807a3cec74c1434fb741a1d1c779d92dff532e253d8b0269", "generation_seconds_batch": 16.483289979980327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1573801040649414, "target_z": -0.7871092739872925, "ce": 6.332377910614014, "genuineness": 4.13078498840332, "identity": 0.43654724955558777, "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/casual_not_formal/541757d064ffa7556ce493ab560e49e4cee28d666e7b6a05bc30d0c14335945e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "544e65c7a7f46b07dcff7486aa67c45278f3bfdd235c9a7a9da5d2988ff1f9ee", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "54acc59f391fc116934946a5e4d62af831a37b72bcafbc52bc8e4b66b03142e6", "generation_seconds_batch": 23.33904941997025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3542470932006836, "target_z": -0.44433597484031895, "ce": 5.974957466125488, "genuineness": 4.3504462242126465, "identity": 0.574626088142395, "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/casual_not_formal/544e65c7a7f46b07dcff7486aa67c45278f3bfdd235c9a7a9da5d2988ff1f9ee.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "d0a2e15321eb95fe5c2049cae42a00b89e7b7fbb0b84787088e5c6ebf54198b5", "generation_seconds_batch": 15.921528177917935, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.098996877670288, "target_z": -2.195312524894611, "ce": 6.205295085906982, "genuineness": 3.8922953605651855, "identity": 0.6017537117004395, "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/casual_not_formal/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "54879ad5e61979945adbf6cec6f78943e9ae4958e33f3a61bd6d27b6a229bbe5", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c74898803b8b9a89bb8bf086b0ae4222cbb0196369e8c87136f449413d6393ca", "generation_seconds_batch": 16.635965156019665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5941903591156006, "target_z": -1.0898436896955561, "ce": 6.292664527893066, "genuineness": 4.425758361816406, "identity": 0.5574430823326111, "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/casual_not_formal/54879ad5e61979945adbf6cec6f78943e9ae4958e33f3a61bd6d27b6a229bbe5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "55c572dabb636f008970229a4552d41f33a77658ba8e5d41b090aa8de8230893", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c1416dc67eb8efbc9f7fc75ed6d0a97a86a57a90460dc9e1cc3d152b6f4b0b0", "generation_seconds_batch": 24.102148448000662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.706240653991699, "target_z": -0.13891599971670782, "ce": 6.022613525390625, "genuineness": 4.507347106933594, "identity": 0.596092700958252, "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/casual_not_formal/55c572dabb636f008970229a4552d41f33a77658ba8e5d41b090aa8de8230893.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "55cfea4c184f42f5b3076e1c3b2ba4d7ccb9eceb587de30ce324e41485b36f98", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bd448da5eb6b151ef8953df0c8934d0a99e595f39491d4c30b2558cbd55f501", "generation_seconds_batch": 16.736226157983765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -1.1250000419520299, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.5532941818237305, "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/casual_not_formal/55cfea4c184f42f5b3076e1c3b2ba4d7ccb9eceb587de30ce324e41485b36f98.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "55e9ecd202974ae9387a8359db30326bd1e2e337a85f7e5586e04e7e1699e7be", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43c300b5fa3ba56cca6c0d8307c006a223757475396c02f1469a0d7a0451c213", "generation_seconds_batch": 18.265823861933313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968777656555176, "target_z": -0.09008795153372351, "ce": 6.109982967376709, "genuineness": 4.419482707977295, "identity": 0.6286442875862122, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre.", "wer": 0.3684210526315789, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/55e9ecd202974ae9387a8359db30326bd1e2e337a85f7e5586e04e7e1699e7be.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "55ea98fe1539045994a5dd718a86ce7abc01112add9ae81c11b58f39e21630bb", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab3e630bfc000dcca699f7106c86acc2fc0a9f2155b935dbc9c9d07b0aa1e888", "generation_seconds_batch": 17.114031622069888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.87827730178833, "target_z": -0.6738280741014656, "ce": 6.133810997009277, "genuineness": 4.256305694580078, "identity": 0.7934128046035767, "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/casual_not_formal/55ea98fe1539045994a5dd718a86ce7abc01112add9ae81c11b58f39e21630bb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "562551345d4661361511e0dff1c17e4eb1bc7a0f3e3ca07fb7b0b51b4d4281a8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddd3edd53ade38a365b46c18d274359644c542fb2a3fda40419486d5fb869a1c", "generation_seconds_batch": 21.837526100920513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7038378715515137, "target_z": -0.9804687599725728, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.5863194465637207, "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/casual_not_formal/562551345d4661361511e0dff1c17e4eb1bc7a0f3e3ca07fb7b0b51b4d4281a8.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": "5692a1557c5d572444f87df13a6e47a356c91c0efde4ed1641cbd7e274c0f56b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "731e230a021dcf53bbdcba3e69bf29161da1cb1620acb1470153c7b7c2edf047", "generation_seconds_batch": 17.292070416966453, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.047732353210449, "target_z": -0.6250000186605564, "ce": 6.173524379730225, "genuineness": 4.306513786315918, "identity": 0.7085450887680054, "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/casual_not_formal/5692a1557c5d572444f87df13a6e47a356c91c0efde4ed1641cbd7e274c0f56b.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": "56a1f5f492e95173d9a5fc29b0bc2535d5a3bb179c5af0a1f8c3d74be3375436", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "fefc47bd5102530ff26056cc7cd675d6859dfc57d3d64910cfeb0fe43fd6bdeb", "generation_seconds_batch": 18.7826644720044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.501274585723877, "target_z": -0.4062499626308688, "ce": 6.316492557525635, "genuineness": 4.137060642242432, "identity": 0.25211101770401, "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/casual_not_formal/56a1f5f492e95173d9a5fc29b0bc2535d5a3bb179c5af0a1f8c3d74be3375436.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": "56cbd6c494d071d0940583276ce3d84b64e07845cf4ba32fd84081b2795bda9e", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "70956dce03dbccf9b604633023c351e1614de7a111f992f130f562a41d7b9e2d", "generation_seconds_batch": 18.350836622994393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.356739044189453, "target_z": -0.5605469329369615, "ce": 6.340320587158203, "genuineness": 4.174716949462891, "identity": 0.2405983805656433, "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/casual_not_formal/56cbd6c494d071d0940583276ce3d84b64e07845cf4ba32fd84081b2795bda9e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "56cdbf35292f4ffd891a5ec05916bd24240726a8746e30e69d8b1bea3287756c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "0965e2d254e8b661523ae3322440645d435b4f63a147009f44c8578b2815c671", "generation_seconds_batch": 15.516184744075872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.604158401489258, "target_z": -0.9257812848671834, "ce": 6.086154937744141, "genuineness": 4.143336772918701, "identity": 0.7015584111213684, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/56cdbf35292f4ffd891a5ec05916bd24240726a8746e30e69d8b1bea3287756c.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": "56f9b446977f4916da82f3bfca7235a57bb76ea6af774afcf33d5f13585cefc5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9175724b0818bc63247185ff8c02a21b2c5ce11acb6b55dae24c0ff9ed1fa1e", "generation_seconds_batch": 21.88863990199752, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5797722339630127, "target_z": -0.1855468852260877, "ce": 6.324435234069824, "genuineness": 4.444586753845215, "identity": 0.2281402200460434, "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/casual_not_formal/56f9b446977f4916da82f3bfca7235a57bb76ea6af774afcf33d5f13585cefc5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "57884bb1dd1c1769bbf61f9334bcf7061895dea8afddf11a4fd2544840d0c14c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff9af11ab3c35e46dfbd42750942daaf5c8c9cbdcabafb7aa0620311ced094f9", "generation_seconds_batch": 21.491997733945027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.259551525115967, "target_z": -0.4404297202710239, "ce": 5.967014789581299, "genuineness": 4.3881025314331055, "identity": 0.6657803058624268, "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/casual_not_formal/57884bb1dd1c1769bbf61f9334bcf7061895dea8afddf11a4fd2544840d0c14c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "57cb529fde3d3fc9af351a520c3e9f0b95f3b2e8daeb0b3860927a90e44da5bf", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e967ba6cfb1ced4680ce93962a2bf18d0cd77182f44166207ca82ae2c48fc21", "generation_seconds_batch": 23.079550773021765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.187499979350536, "ce": 6.268836498260498, "genuineness": 4.3504462242126465, "identity": 0.5119212865829468, "hypothesis": "De deur staan nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/57cb529fde3d3fc9af351a520c3e9f0b95f3b2e8daeb0b3860927a90e44da5bf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "581b3c70efa4d5938d318157860c4f3e90185df115a6b6225e65c40f67540a5a", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5163c3640e1c4a0ce3b81be8e6bd01dce7c5d3ef56a6012a51bbf303282bdce2", "generation_seconds_batch": 15.955347063019872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.2265625575335737, "ce": 6.419747352600098, "genuineness": 4.507347106933594, "identity": 0.5180035829544067, "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/casual_not_formal/581b3c70efa4d5938d318157860c4f3e90185df115a6b6225e65c40f67540a5a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "584e47cf6b5f319adaf99f10479b17013525ce56b9288c6252b942062fedf674", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9e122d550d026e563f3cc9ff08e54de44180ae02c7662836eda6e514e12a3dd", "generation_seconds_batch": 17.831452472950332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -0.8984374228270204, "ce": 5.871703147888184, "genuineness": 4.262581825256348, "identity": 0.6732572913169861, "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/casual_not_formal/584e47cf6b5f319adaf99f10479b17013525ce56b9288c6252b942062fedf674.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "5875372666f426be98dc237268979e69b3b8915b9cb6de2023f87c51df5a89dc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ed6285bc50714f2e73307af4af2a7ead6747757c53692fc8bb9e231955d602de", "generation_seconds_batch": 24.592695421888493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3168673515319824, "target_z": -0.5429687257512837, "ce": 6.229123115539551, "genuineness": 4.469690799713135, "identity": 0.7311526536941528, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise, dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/5875372666f426be98dc237268979e69b3b8915b9cb6de2023f87c51df5a89dc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "5878b6b8b55e50838cbe99b0955cb35eacde87d92d5269dcfd8add45c32e89ea", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0a9a837c47211806d0da06b00f4073a6c954e187cd78d7c3a5c8f09520ff9355", "generation_seconds_batch": 17.18123829201795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.305119514465332, "target_z": -1.1445313635368497, "ce": 6.173524379730225, "genuineness": 4.482243061065674, "identity": 0.579508364200592, "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/casual_not_formal/5878b6b8b55e50838cbe99b0955cb35eacde87d92d5269dcfd8add45c32e89ea.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": "58b831230a6ecec8823f13db8bd0d4437a511174a6961b538077fcbc413dd234", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e3503f8bacd8259dc12de6f784f4cc91488c3a997727e120882aedd20eca0790", "generation_seconds_batch": 20.70403451297898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.391627073287964, "target_z": -0.4648437522650376, "ce": 5.959072113037109, "genuineness": 4.369274139404297, "identity": 0.6672972440719604, "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/casual_not_formal/58b831230a6ecec8823f13db8bd0d4437a511174a6961b538077fcbc413dd234.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "58d5a4e771c5238355e34d222289ecaf04296126f2020c1ae33adb946bc93719", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df8e03255a3e5173d8e15b6deb89d62b9ca478b36851ed46ef30fe5cbefe2d83", "generation_seconds_batch": 25.2426143969642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4265148639678955, "target_z": -0.3222656391632929, "ce": 6.149696350097656, "genuineness": 4.582659721374512, "identity": 0.6118761301040649, "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/casual_not_formal/58d5a4e771c5238355e34d222289ecaf04296126f2020c1ae33adb946bc93719.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "58ed1bfe8f207541d2d87b381daec3a4e9fe47c6747db612f0cccee959ead8b1", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7a63ace64b2d3ba465076588931be685d136fd5b7f7c6d88fcf98c38b0b93dad", "generation_seconds_batch": 19.162483896012418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1755361557006836, "target_z": -1.2851562791829787, "ce": 6.006728172302246, "genuineness": 4.250029563903809, "identity": 0.6487685441970825, "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/casual_not_formal/58ed1bfe8f207541d2d87b381daec3a4e9fe47c6747db612f0cccee959ead8b1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "598270f42a7076f0255e48f77dc123eb05236497cc62968ec540b43b9be5734a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a75d7c2314d1d29a9319e6d9d518819f9ac4e3eb9e22ad151ba800f55bb1082c", "generation_seconds_batch": 21.911721925018355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8065433502197266, "target_z": -0.10302727920280887, "ce": 6.268836498260498, "genuineness": 4.394378662109375, "identity": 0.6423650979995728, "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/casual_not_formal/598270f42a7076f0255e48f77dc123eb05236497cc62968ec540b43b9be5734a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "5a4947eb9a29ce2e59c8e2aa805ef9d14233b843f098cd3e7e1d1a06360e25da", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2429dcf84bd2dcdca8c30f1e42bf11f84e4c36b87be84a9130cdb579980c3ae3", "generation_seconds_batch": 25.03010569489561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0975723266601562, "target_z": -0.5722656019734433, "ce": 6.284721851348877, "genuineness": 4.042920112609863, "identity": 0.265980064868927, "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/casual_not_formal/5a4947eb9a29ce2e59c8e2aa805ef9d14233b843f098cd3e7e1d1a06360e25da.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "5aaf54e03a449d09bf71cb7b3a72cad6a1c3e100b4bf177aaf7dd2096cddce28", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd9ff578fab45bf842cec4e8d9c6c583cb11fbeaaa24de9ea1608fb92fcfd812", "generation_seconds_batch": 22.544980152044445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4379072189331055, "target_z": -2.0468749440447507, "ce": 6.340320587158203, "genuineness": 3.9425034523010254, "identity": 0.517340898513794, "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/casual_not_formal/5aaf54e03a449d09bf71cb7b3a72cad6a1c3e100b4bf177aaf7dd2096cddce28.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": "5acfda3a98118a1e7cbd30faed13dd5060c26bc34d19023524c0f22b439d7c9d", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b02e97d8398437c6d528ff3799cb62042e3ea3e1f7d37194fda13fc54c1f71ef", "generation_seconds_batch": 22.449216627981514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4040870666503906, "target_z": -0.47363279429885474, "ce": 6.395919322967529, "genuineness": 4.231201171875, "identity": 0.49912524223327637, "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/casual_not_formal/5acfda3a98118a1e7cbd30faed13dd5060c26bc34d19023524c0f22b439d7c9d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "5ad43e064c0999f117a42f862834b36d9cd89619b50a2425e10ad5e5ce1d4352", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c4faaf5f33584e0102181c5c466115fa389aca3965874d3a613bf135b664e38", "generation_seconds_batch": 23.58782120095566, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8932290077209473, "target_z": -0.8945313356120485, "ce": 6.221180438995361, "genuineness": 4.3504462242126465, "identity": 0.7615447044372559, "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/casual_not_formal/5ad43e064c0999f117a42f862834b36d9cd89619b50a2425e10ad5e5ce1d4352.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "5b688e8f2a5318fdb55816114bf4014e1277ce26e6b66f9ff8a94c7371547f98", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84dab4a1f40843b436a42a08a2aac3c6f4a59d9fed0b89da675d75c07a259520", "generation_seconds_batch": 22.74002688098699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.185504198074341, "target_z": -1.2773436452960008, "ce": 6.260893821716309, "genuineness": 3.986435890197754, "identity": 0.419481098651886, "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/casual_not_formal/5b688e8f2a5318fdb55816114bf4014e1277ce26e6b66f9ff8a94c7371547f98.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "5b715cd0278be68167154b87234b4363e78735005f07a77be25bf81750e2189e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8780d8e2ddd1f187a4d206f73c30bef9f0ed6e603ee467f632b593ba3287fd98", "generation_seconds_batch": 16.743776149931364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.7343749958156798, "ce": 6.133810997009277, "genuineness": 4.507347106933594, "identity": 0.7957454919815063, "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/casual_not_formal/5b715cd0278be68167154b87234b4363e78735005f07a77be25bf81750e2189e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "5b88553e9cd42456244ddd48572539152c287447a835fc8535757c0ecd73bf12", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d36cf827b2e6e239d9da47e573778269fa1c0c255715538984e1b941b7d8a13c", "generation_seconds_batch": 18.959574069012888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.878187894821167, "target_z": -0.062255881407501094, "ce": 6.24500846862793, "genuineness": 4.444586753845215, "identity": 0.40553009510040283, "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/casual_not_formal/5b88553e9cd42456244ddd48572539152c287447a835fc8535757c0ecd73bf12.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "5be5ecff49cf4d4fb938ee4fd96299b145aaa7b6a356eb21fe4f53d2f3c9b0e5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db1d2359a0d0905923ecc37b1ad7f9acda08e587697b393a91dd465f77239402", "generation_seconds_batch": 16.392561442917213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.195472240447998, "target_z": -1.1484374507518216, "ce": 6.006728172302246, "genuineness": 4.061748504638672, "identity": 0.3457338809967041, "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/casual_not_formal/5be5ecff49cf4d4fb938ee4fd96299b145aaa7b6a356eb21fe4f53d2f3c9b0e5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "5cc3a7d524a684f709959fa9548ae75321a86fee68ca922be6189b820773c44f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a1bbeee1101e600a5f28c21cf23b09a025e7781c5789abf3d90adf9c6aa0921", "generation_seconds_batch": 23.06769301602617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.967988967895508, "target_z": -0.5703124420334248, "ce": 5.831989288330078, "genuineness": 3.9048471450805664, "identity": 0.5420495867729187, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás abriendo.", "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/casual_not_formal/5cc3a7d524a684f709959fa9548ae75321a86fee68ca922be6189b820773c44f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "5ccdbf17af39b570cf3ce6fbf1417de80c71ba52a61ef1225ee7a4715b5427c5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "592621ca29eecd88460bafd281d22e7e5e9a1a131a53c7ac008906fb40d9c2e5", "generation_seconds_batch": 22.63391266600229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.928117036819458, "target_z": -0.7558593615298244, "ce": 5.732706069946289, "genuineness": 4.262581825256348, "identity": 0.6370294690132141, "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/casual_not_formal/5ccdbf17af39b570cf3ce6fbf1417de80c71ba52a61ef1225ee7a4715b5427c5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "5ce153c04941d349c30dd863748c6c04d04cb1d521aef19c2cec3c9a9fcc223b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "623ab4116f96c4413b6515dc9fee17b5329ee31911383276c6c3624e7c8c37b2", "generation_seconds_batch": 24.657570970011875, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9630048274993896, "target_z": -0.4765625215186649, "ce": 6.149696350097656, "genuineness": 4.068024158477783, "identity": 0.4616590142250061, "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": 0.120000000000001, "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/casual_not_formal/5ce153c04941d349c30dd863748c6c04d04cb1d521aef19c2cec3c9a9fcc223b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "5cfe1f372f86ba3ecf11fe5e3160379efd425d07646c56fe3000da300303c4a8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abdf6954c10f1ab37a1601409e2c7ef5755bd49ef777ac13a200de9ac05265d1", "generation_seconds_batch": 23.67371809692122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121779918670654, "target_z": 0.042480443106444415, "ce": 6.086154937744141, "genuineness": 4.6203155517578125, "identity": 0.6076183319091797, "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/casual_not_formal/5cfe1f372f86ba3ecf11fe5e3160379efd425d07646c56fe3000da300303c4a8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "5d3c8591af76da609435e0d78b0e943d9f913681eba8ee8fa816c3154e602220", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "37a0c49fa485880c6dc1d8bf117d17ec51331c4eb7bc728b10bc9156df90c656", "generation_seconds_batch": 23.831461454974487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9761770963668823, "target_z": -1.6015624862208826, "ce": 5.927301406860352, "genuineness": 3.998987913131714, "identity": 0.5464476346969604, "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/casual_not_formal/5d3c8591af76da609435e0d78b0e943d9f913681eba8ee8fa816c3154e602220.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": "5d3f403530c51697376b9810b635b2f589daf8aae5aa86afb06b1d80777d1431", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcbe9eaf0fa446253b873f2bde7f625d2cd21ae6966e78793217e09f17af930d", "generation_seconds_batch": 26.220426419051364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7984442710876465, "target_z": -0.21166990090738838, "ce": 6.046441555023193, "genuineness": 4.519899368286133, "identity": 0.7175894975662231, "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/casual_not_formal/5d3f403530c51697376b9810b635b2f589daf8aae5aa86afb06b1d80777d1431.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "5d599e8de260ef9e91a1236f9588c60a2d88a0ef4959572cd0c3902c11f3508d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7d328ec9e922cc58f03018149dd9b61fbdaf2768d91f4c23e4b774a64274377", "generation_seconds_batch": 17.71769471198786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6847472190856934, "target_z": -0.03784181398099265, "ce": 6.046441555023193, "genuineness": 3.9550557136535645, "identity": 0.36073699593544006, "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/casual_not_formal/5d599e8de260ef9e91a1236f9588c60a2d88a0ef4959572cd0c3902c11f3508d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "5d95bea7c594645ab1cf24165f937361df40bc390e15e2b2a12723d9e241d1ce", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10e9818c11d23d252c46a4bbafb520ca032fc0ff37bdd55806a643780e6cc506", "generation_seconds_batch": 16.973842874052934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4945108890533447, "target_z": -0.8593750449454032, "ce": 6.292664527893066, "genuineness": 4.262581825256348, "identity": 0.34603190422058105, "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/casual_not_formal/5d95bea7c594645ab1cf24165f937361df40bc390e15e2b2a12723d9e241d1ce.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "5db50d36a4af85324db33bc936f1fcbd3b8944dadb8b1cfc2e4d1b8fa3a5bc24", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ada1c78b4c11e18dd490da9b8199070945c66b76cd1f90b5775ae1e35ad6b757", "generation_seconds_batch": 22.776898788986728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.119287967681885, "target_z": 0.11938476721663019, "ce": 6.157639026641846, "genuineness": 4.532451152801514, "identity": 0.7101927995681763, "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/casual_not_formal/5db50d36a4af85324db33bc936f1fcbd3b8944dadb8b1cfc2e4d1b8fa3a5bc24.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "5de044c73c40e4cac1f270cd73b2d916de6e95aa7396683f02a225b6438b995f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7005bcbcac40905fcac1fef7d5da3d3505ff119f4a5860b5113fda90a00d5adf", "generation_seconds_batch": 22.08326040999964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.295151710510254, "target_z": -1.2773438153243457, "ce": 6.443574905395508, "genuineness": 4.381826400756836, "identity": 0.6641733050346375, "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/casual_not_formal/5de044c73c40e4cac1f270cd73b2d916de6e95aa7396683f02a225b6438b995f.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "3983804095b381c2ed3f04d33a2f7cf698a5462aff874bbd6f385d89e0c528d8", "generation_seconds_batch": 16.339050746057183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0658886432647705, "target_z": -1.4570312827958345, "ce": 6.086154937744141, "genuineness": 4.243753433227539, "identity": 0.7576144337654114, "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/casual_not_formal/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "5e4131b86f5565f281639dff77e24dcf11f619a9017668a02d5c46abf8c5feaa", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a09b3456fd1233119108e660c2fadfd361c25783942ab6a7196f1c52d4fa22b", "generation_seconds_batch": 24.258394913980737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.383439064025879, "target_z": 0.36914071317642577, "ce": 6.038498878479004, "genuineness": 4.3881025314331055, "identity": 0.6376621723175049, "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/casual_not_formal/5e4131b86f5565f281639dff77e24dcf11f619a9017668a02d5c46abf8c5feaa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "5e6e4deb4a4e6e9dea12fb054a9db41c433569e056d6de7303358deb8242d6d0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63a7aee7f0d1ab6b08a867462a105a698c27321ffa1f54f72bc3df2c954b4f26", "generation_seconds_batch": 22.964893874013796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.22395133972168, "target_z": 0.1796874688409138, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.38305404782295227, "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/casual_not_formal/5e6e4deb4a4e6e9dea12fb054a9db41c433569e056d6de7303358deb8242d6d0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "5e7c0af10af21514dd69520500ba514380f8416682f59259cce9958803620388", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49658354ce7258555eaa6ae06aa34d4105076da33b3e9a670f057c9e459fe556", "generation_seconds_batch": 17.545386334997602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.391627073287964, "target_z": -0.34179683893466617, "ce": 6.133810997009277, "genuineness": 4.432034492492676, "identity": 0.6001226902008057, "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/casual_not_formal/5e7c0af10af21514dd69520500ba514380f8416682f59259cce9958803620388.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "5e87228b0e14d118d9590c556ce87aae9d26973bda5e61da5ee7d53d6679d647", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f6c3a39c60f67308bf9efb189e2a57840347321886e02d5ca1a57fc661aeef54", "generation_seconds_batch": 22.362256914027967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.154888153076172, "target_z": -0.7998046370096523, "ce": 6.173524379730225, "genuineness": 4.394378662109375, "identity": 0.6401684284210205, "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/casual_not_formal/5e87228b0e14d118d9590c556ce87aae9d26973bda5e61da5ee7d53d6679d647.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "5ed9ff930986f55e2618500fc858690c005cc82e0678f87eac5b0087c565701a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bff2b54b64102110cbd104bc5797f3c7e61b405d52546e44c732f11c01813982", "generation_seconds_batch": 25.48835280793719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8166897296905518, "target_z": -1.7460937398849627, "ce": 6.435632705688477, "genuineness": 4.024092197418213, "identity": 0.5610401630401611, "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/casual_not_formal/5ed9ff930986f55e2618500fc858690c005cc82e0678f87eac5b0087c565701a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "5f55abd6fc7cb288d7c79c5c9a369b9eba4018080ba7f3d6539989b8a5066306", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "379e7c2722b00869cd62f05723bfe278cba3fecc57966d37b7a3622352728619", "generation_seconds_batch": 26.02787971496582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5710501670837402, "target_z": -0.34570316638297227, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.7144823670387268, "hypothesis": "Je trouvai la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle 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/casual_not_formal/5f55abd6fc7cb288d7c79c5c9a369b9eba4018080ba7f3d6539989b8a5066306.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "5f58908e1c0d673f00f7710e5f2920353d1de8aee1f8118024466e628df0d677", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a87a62b020af0e96ae33554943bd4135ee231b5bc8740ecaf24d7f0d1ba15d59", "generation_seconds_batch": 17.575322061078623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.768629550933838, "target_z": -0.8671875087392081, "ce": 6.2767791748046875, "genuineness": 4.519899368286133, "identity": 0.730095386505127, "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/casual_not_formal/5f58908e1c0d673f00f7710e5f2920353d1de8aee1f8118024466e628df0d677.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "5f8c4964bee5926b2f01c589ccfceec9fed7f52c70869c30495ce7f473bfc65d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d5c0f33a1fca894b6d4fe5143f4e6dc0bc54bfc1bdb9def5c57d7cf0c400f44", "generation_seconds_batch": 20.984149588039145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.096859931945801, "target_z": 0.10888666926111182, "ce": 6.09409761428833, "genuineness": 4.406930446624756, "identity": 0.6641795635223389, "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/casual_not_formal/5f8c4964bee5926b2f01c589ccfceec9fed7f52c70869c30495ce7f473bfc65d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "601b2bbf71af5050249f94491e78eb81621aa9cec096895845cf9bffb234f497", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f9e38b9e660ffaf18ed4937e30bf40b2e9fe3ac920fae81955de5b897b8a18a", "generation_seconds_batch": 22.623998783063143, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5909862518310547, "target_z": -0.3085937383514589, "ce": 6.09409761428833, "genuineness": 4.444586753845215, "identity": 0.7696021795272827, "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/casual_not_formal/601b2bbf71af5050249f94491e78eb81621aa9cec096895845cf9bffb234f497.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "6054889a537a8022e5b295e8fdadfdd97ffd7b8aa4cbf529feea0566b254b57f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f14b68fc68d644ccac8d3cf1b3a46836495c48610378df382b65bb1d444a455", "generation_seconds_batch": 24.503014539950527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.7929687164214073, "ce": 6.252951145172119, "genuineness": 4.1998209953308105, "identity": 0.3835742473602295, "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/casual_not_formal/6054889a537a8022e5b295e8fdadfdd97ffd7b8aa4cbf529feea0566b254b57f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "60a3f14415d4d892c94482009f41bbed74815775e1c774e639bc03e9aeb1e059", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7366770ec15c69bb7c1b770b56c97a94b03d2f8b4c2790bad7477ac3eaedb730", "generation_seconds_batch": 22.858587356982753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.047732353210449, "target_z": -0.7558594560051364, "ce": 6.046441555023193, "genuineness": 4.65797233581543, "identity": 0.8179250955581665, "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/casual_not_formal/60a3f14415d4d892c94482009f41bbed74815775e1c774e639bc03e9aeb1e059.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "60b183bd7b0e3534916800196672771b08e6c3ad70d071a7f3f5c81b23b69ebf", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "abfc1e8b52da4d35c5c16b83fee9d2fa29e6d74ddd32b7113bc24ba039c9cbe0", "generation_seconds_batch": 16.54903685906902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.214695692062378, "target_z": -0.6005859301638383, "ce": 6.070269584655762, "genuineness": 4.65797233581543, "identity": 0.6377583742141724, "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/casual_not_formal/60b183bd7b0e3534916800196672771b08e6c3ad70d071a7f3f5c81b23b69ebf.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": "60ea657951708746ebcabb27d14142490917a1f5f4b4ce80e68ed92345825726", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c4e5ad69c91037d7f9ea0b002b36eb4cd73499cc5ec06f9c46db6d1e064fcbe8", "generation_seconds_batch": 26.167363791028038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6970741748809814, "target_z": -1.8515626061414583, "ce": 6.252951145172119, "genuineness": 3.9801597595214844, "identity": 0.6080907583236694, "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/casual_not_formal/60ea657951708746ebcabb27d14142490917a1f5f4b4ce80e68ed92345825726.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "60f8d40d2ea21c90e2ecbe10fc11dff9df985356d36d5454ff3fabbae1468266", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a836f3b5aff45627ea2a246cba3d8b850f1e7891c02ed6ebba5877c756b311d", "generation_seconds_batch": 24.447385976091027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.1953123589113197, "ce": 5.816103935241699, "genuineness": 4.344170093536377, "identity": 0.5593946576118469, "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/casual_not_formal/60f8d40d2ea21c90e2ecbe10fc11dff9df985356d36d5454ff3fabbae1468266.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "61111ef5e7421630d96d984e15f7dd73e75b07057e1c1801b02794ae6c025273", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3237a63e0ffade2fd82b86252222b7df783b2259f252c640adcd61854a620b33", "generation_seconds_batch": 18.1723309190711, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.047732353210449, "target_z": -0.5527344227546938, "ce": 6.292664527893066, "genuineness": 4.231201171875, "identity": 0.3763004541397095, "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/casual_not_formal/61111ef5e7421630d96d984e15f7dd73e75b07057e1c1801b02794ae6c025273.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": "61278478a85b38b21624a56b48bc2b1cc726d14d7c8b8b2ff4ef26dc35a6aa36", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afd3024e6ceba7220cd621dd2b515586f69d350c48d69e65b348e756f8fcc87b", "generation_seconds_batch": 16.13566301204264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.468878746032715, "target_z": -0.3525390781631796, "ce": 6.475345611572266, "genuineness": 4.494794845581055, "identity": 0.5895187854766846, "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/casual_not_formal/61278478a85b38b21624a56b48bc2b1cc726d14d7c8b8b2ff4ef26dc35a6aa36.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "61b6e403ba27b1f326499efa6fc2e31b6822cd2e0c442febbd829d866d9ae0ce", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1a8f6be4e3584123eccd4f4737f2278b81e638a75a1ca2577511b413f6f9426", "generation_seconds_batch": 24.989573237020522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.139224052429199, "target_z": 0.11840828106963455, "ce": 6.23706579208374, "genuineness": 4.532451152801514, "identity": 0.6063778400421143, "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/casual_not_formal/61b6e403ba27b1f326499efa6fc2e31b6822cd2e0c442febbd829d866d9ae0ce.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "61e51b872bcb2a263d2a9e85cfa380203cebd0f12513f721683aecae3d242d86", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e5a47ef1d025448f6f034bf937802a5e84f4df58272995475a9c8f154d5eeb5", "generation_seconds_batch": 16.571312663028948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5475549697875977, "target_z": -1.9179687586343566, "ce": 6.475345611572266, "genuineness": 4.105680465698242, "identity": 0.3953155279159546, "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/casual_not_formal/61e51b872bcb2a263d2a9e85cfa380203cebd0f12513f721683aecae3d242d86.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "61f87d28f1d1eb3336992f2a488c4f982688b9ffee6e49bb62ec36feb295edcc", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a6e53a7143bea4f0707a86e874bc5334fa60bdc977eeb49259fea0a1f56b3a7", "generation_seconds_batch": 18.18650393805001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9672765731811523, "target_z": -0.007324204897449768, "ce": 6.260893821716309, "genuineness": 4.557555198669434, "identity": 0.27607378363609314, "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/casual_not_formal/61f87d28f1d1eb3336992f2a488c4f982688b9ffee6e49bb62ec36feb295edcc.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": "6209bc83e09ddb8202d272edfbe6dad095eff88cc1f52de7b8bb6c78cd13931f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "36298f3a30e8395368fe8983a9ef6d525b73b78af09db7b25a2b1ab7510d572a", "generation_seconds_batch": 17.378680405090563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7486937046051025, "target_z": -1.0800781903195447, "ce": 6.0146708488464355, "genuineness": 4.519899368286133, "identity": 0.6889193058013916, "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/casual_not_formal/6209bc83e09ddb8202d272edfbe6dad095eff88cc1f52de7b8bb6c78cd13931f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "621040314aeaf5b53cc1fd4345b09fa99f2600761d0e02bf3e407bcdbb954231", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "388458b5ad9e1a084dbab60d3ebc117cc2f2d543fab1a61ab0ddaa1af073713d", "generation_seconds_batch": 22.268308773986064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.802805185317993, "target_z": 0.09326169593814118, "ce": 5.990842819213867, "genuineness": 4.300237655639648, "identity": 0.816236674785614, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen haar samen.", "wer": 0.16666666666666666, "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/casual_not_formal/621040314aeaf5b53cc1fd4345b09fa99f2600761d0e02bf3e407bcdbb954231.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "622eac44c7b1c6d8bde43ad3878919f53071e735dd238c16246c8e1ece0f5c9b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dc4c90e068a6f43ef5c4674771c2e8806f42996c9be245a8e7ef2779a289a6f", "generation_seconds_batch": 22.76991375302896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.499494791030884, "target_z": -1.2089844075685305, "ce": 6.157639026641846, "genuineness": 4.557555198669434, "identity": 0.6935238838195801, "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/casual_not_formal/622eac44c7b1c6d8bde43ad3878919f53071e735dd238c16246c8e1ece0f5c9b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "624891db993a76eba9a5adc2d69a6e53b7dfe576ab6cf7b04bbfccf4a221b249", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "620806bd4c3d956411eaff4e90d3e0cf498bb4cdea58cf0ff15fbfeec1f7d393", "generation_seconds_batch": 24.595484151039273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.109320163726807, "target_z": 0.21435552230635424, "ce": 5.816103935241699, "genuineness": 4.344170093536377, "identity": 0.5471473932266235, "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/casual_not_formal/624891db993a76eba9a5adc2d69a6e53b7dfe576ab6cf7b04bbfccf4a221b249.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "625a6c9a6a027beef8518d7624f3613726c05585621d197c63fdfa83d55559cb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd83041edd0d00778e7de58d52f16690f2b380a5739f6ac370a543d00c3daff9", "generation_seconds_batch": 16.491749526001513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.115016222000122, "target_z": -0.3759765749785201, "ce": 6.260893821716309, "genuineness": 4.469690799713135, "identity": 0.19673846662044525, "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/casual_not_formal/625a6c9a6a027beef8518d7624f3613726c05585621d197c63fdfa83d55559cb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "626913f1cc11027cf6872aaaf614b89b6ca698b894c3b176c882272895309f8c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cc082e2541c8d2eb44fc0f3c1aca5780466e85727795e04f838cb2d172ffa867", "generation_seconds_batch": 23.908204993000254, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.777173042297363, "target_z": 0.630859317163085, "ce": 6.141753673553467, "genuineness": 4.406930446624756, "identity": 0.4902650713920593, "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/casual_not_formal/626913f1cc11027cf6872aaaf614b89b6ca698b894c3b176c882272895309f8c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "629d6ea09f5d258011d43441941af98930f5d3bbd2a6d39255d7fcf2fdaa46e7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "accb7f59b0b020f5c3fa5f49e2cb6279d91489ba2f4f2851ca4240f998d6ddd6", "generation_seconds_batch": 22.944294171989895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5742545127868652, "target_z": -0.9765624838069766, "ce": 6.300607204437256, "genuineness": 4.306513786315918, "identity": 0.3839026689529419, "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/casual_not_formal/629d6ea09f5d258011d43441941af98930f5d3bbd2a6d39255d7fcf2fdaa46e7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "631de1a37ca15a5cc7ac09a654c55eb8a74e8b4a0aeb9a186e9d7d793cf3bfac", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b3418524f9088c8f777cbce7a3796f6d31ca60bd66dfdede01355617ce9344c5", "generation_seconds_batch": 23.079550773021765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5212104320526123, "target_z": -0.3183593697183794, "ce": 6.133810997009277, "genuineness": 4.319066047668457, "identity": 0.6806759834289551, "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/casual_not_formal/631de1a37ca15a5cc7ac09a654c55eb8a74e8b4a0aeb9a186e9d7d793cf3bfac.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "63216da3f402fb58fc5ca57fc184fec7d9171cc44e2250fc2236ecaf11b25379", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bce3147758c0b8b447ced1422a4d64f6bfcf07813246bb26536755a113ee3996", "generation_seconds_batch": 17.72676364099607, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9131650924682617, "target_z": -0.7167968891080669, "ce": 6.252951145172119, "genuineness": 4.570107460021973, "identity": 0.6209117770195007, "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/casual_not_formal/63216da3f402fb58fc5ca57fc184fec7d9171cc44e2250fc2236ecaf11b25379.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "633ff5b2d56fecd10d6fe84fe1fae1fd1e7ca9bed9187b88b7357a8241948506", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9582d1ca65b334aa4aadcea35b25b974d8028903077b5078b5523dd91c9ca3c9", "generation_seconds_batch": 23.360062603023835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6177749633789062, "target_z": 0.011962874388955533, "ce": 5.796247482299805, "genuineness": 4.218649387359619, "identity": 0.06091739982366562, "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/casual_not_formal/633ff5b2d56fecd10d6fe84fe1fae1fd1e7ca9bed9187b88b7357a8241948506.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": "636eafaa7a7e66a53ed64ccef42dcdaa9fd1f245b80dc36ae45dd796b989fe42", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55a4cf0769c49f1d1abe0091caed8b68a1cede4d875efdd36fe38428d6823ba9", "generation_seconds_batch": 17.378680405090563, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8594980239868164, "target_z": -0.21508786015783338, "ce": 6.109982967376709, "genuineness": 4.65797233581543, "identity": 0.5955688953399658, "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/casual_not_formal/636eafaa7a7e66a53ed64ccef42dcdaa9fd1f245b80dc36ae45dd796b989fe42.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "63c9eac8a708a6785ec79d2bfee047f7a86bdbfaf856a4242239cacc22f62f56", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "622ce04a1b2c549b5c0beda0edaa1ffdea43211aa5b88986239083feaf27e3ea", "generation_seconds_batch": 18.906176304910332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9080917835235596, "target_z": 0.12719727889108076, "ce": 6.078212261199951, "genuineness": 4.356721878051758, "identity": 0.5786406397819519, "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/casual_not_formal/63c9eac8a708a6785ec79d2bfee047f7a86bdbfaf856a4242239cacc22f62f56.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": "63fa7246a51fd52b531ab1c2367f5be374adb7672e9832fdaecfafcdbe8f96f2", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e5099020b1c18c45586a4c8130dc5e36710465c74b9626b3b20bd021dfd53416", "generation_seconds_batch": 22.449216627981514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.658892869949341, "target_z": -0.2839355629715793, "ce": 6.41180419921875, "genuineness": 4.507347106933594, "identity": 0.19774286448955536, "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/casual_not_formal/63fa7246a51fd52b531ab1c2367f5be374adb7672e9832fdaecfafcdbe8f96f2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "64205f4f11cecd81ed044d9799543fa7f6c1e477fcceff33312620364ec36a08", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d395813acfddcce801c9d62bb2a255f3832da6937b894f64af0bde7a8c9be4f9", "generation_seconds_batch": 22.106512819998898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.453214645385742, "target_z": 0.6777344227546938, "ce": 5.863759994506836, "genuineness": 4.2751336097717285, "identity": 0.19702911376953125, "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/casual_not_formal/64205f4f11cecd81ed044d9799543fa7f6c1e477fcceff33312620364ec36a08.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": "6487b3a8865bb10742fd5c47ebb5155c93f0a160d27b35d9e9cc0b0a28801723", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8b608a00595ea37541544e21d9ea1df91f83030ed29ad03c321aa7b58a1c8c0", "generation_seconds_batch": 24.893112823017873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9131650924682617, "target_z": -0.8593750238061126, "ce": 5.935244083404541, "genuineness": 4.507347106933594, "identity": 0.8136078119277954, "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/casual_not_formal/6487b3a8865bb10742fd5c47ebb5155c93f0a160d27b35d9e9cc0b0a28801723.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "648f9edb48769f2139ece3879d4a92939e131e52395e8e2da45bda676754d60e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d21d24f57506b9fa9173281df1a100a23b2727f2ab2b24c6c2121b1bf13ad24d", "generation_seconds_batch": 22.385621860041283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7867858409881592, "target_z": -1.6992187728744863, "ce": 6.2767791748046875, "genuineness": 4.21237325668335, "identity": 0.5902172327041626, "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/casual_not_formal/648f9edb48769f2139ece3879d4a92939e131e52395e8e2da45bda676754d60e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "649c7e87a1d7fc1e868ecfd7b6cd1691f8a474084673de63e99956f985943166", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3bc0c6d3e0903328a5d6300255a6c0db4278030a5d46d07aa584592dc3ed8ef", "generation_seconds_batch": 23.37159023899585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6339728832244873, "target_z": -0.0437012242858279, "ce": 6.054384231567383, "genuineness": 4.18099308013916, "identity": 0.562126874923706, "hypothesis": "Nan! La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás lista.", "wer": 0.13333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/649c7e87a1d7fc1e868ecfd7b6cd1691f8a474084673de63e99956f985943166.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "64b2b2828fa5cdb48bc846fc63c360838a4e93b319407c402d4b642f9fed4752", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "986100b4e7508cbbe1c39ee786b74df306b5beca4318bc8e6dc275722689fe30", "generation_seconds_batch": 22.089328057016246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.6328124117275334, "ce": 5.974957466125488, "genuineness": 4.457139015197754, "identity": 0.7259929180145264, "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/casual_not_formal/64b2b2828fa5cdb48bc846fc63c360838a4e93b319407c402d4b642f9fed4752.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "64b809b12dd1b73baf3b20fa5a1236dc0bd7ce6e93c3d6b6a868eb29e2373905", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a985c0111f1263ab35470db77b281710a806471fc0dff0b10f61dd6f162b7234", "generation_seconds_batch": 15.40404078294523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6470558643341064, "target_z": -0.14892582596369883, "ce": 5.9431867599487305, "genuineness": 4.2249250411987305, "identity": 0.6486924886703491, "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/casual_not_formal/64b809b12dd1b73baf3b20fa5a1236dc0bd7ce6e93c3d6b6a868eb29e2373905.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "64ebff0d288237f5af6693fdc845c0c04c1f4a5982f758be15e69d299a7f2894", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c61d6de4199131d7db6a539811b6942242addf5fa5af4aa28749b5f43e7087b", "generation_seconds_batch": 15.27358686097432, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6071841716766357, "target_z": -0.07421874751025768, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.6169546842575073, "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/casual_not_formal/64ebff0d288237f5af6693fdc845c0c04c1f4a5982f758be15e69d299a7f2894.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": "650222a698a5c1bab47fa35ddce04e5f619293eebe18501ededc76704071f783", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "db77eaf93c7812b8bb6ebe12104d54e79ffcec9c61b0742041b02cc80edb1361", "generation_seconds_batch": 18.20180156605784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4845428466796875, "target_z": -0.37353517356373744, "ce": 3.202965259552002, "genuineness": 4.168440818786621, "identity": -0.14721617102622986, "hypothesis": "", "wer": 1.0, "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/casual_not_formal/650222a698a5c1bab47fa35ddce04e5f619293eebe18501ededc76704071f783.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "655bd0dbd7d3579a7232da3a7f483c39cacee9aae56666d4cf1b17e2b76b95d6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "457a74cf6af0dfddfaf3248de2cbe25f07a01199d6823a76fe45f95c1a184189", "generation_seconds_batch": 26.815919438027777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9630048274993896, "target_z": -0.6328125289076522, "ce": 6.197352409362793, "genuineness": 4.494794845581055, "identity": 0.7371729016304016, "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/casual_not_formal/655bd0dbd7d3579a7232da3a7f483c39cacee9aae56666d4cf1b17e2b76b95d6.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": "663539c83be0dd9f9b10802eca74b29f433f69b37a66ecfaeb511afb6af3016f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c162ba23b0d3185000ca0154046b527e54e1d8aefdaf5409271cf381948edd60", "generation_seconds_batch": 23.89727048401255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.129255771636963, "target_z": 0.21093746879209185, "ce": 6.348263263702393, "genuineness": 4.65797233581543, "identity": 0.7471842169761658, "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/casual_not_formal/663539c83be0dd9f9b10802eca74b29f433f69b37a66ecfaeb511afb6af3016f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "6653e923bc2bb0de15eb526939aab7c49ffbe0d9599089faf5247e3d863add13", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e2228200d15f314029366102b19005137efc051a85c4fffd4a8edc76aaabcb0", "generation_seconds_batch": 19.15707281907089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.478846549987793, "target_z": -0.3857422148925589, "ce": 6.038498878479004, "genuineness": 4.65797233581543, "identity": 0.5960589051246643, "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/casual_not_formal/6653e923bc2bb0de15eb526939aab7c49ffbe0d9599089faf5247e3d863add13.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": "66788c625004026c856e345211c2e71b4624ca939eba9f6db5c39bf0e742fabf", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "99589bfa5c05ddcf0e67f6218fc10e1598c6d9bc55e678a74162cf0099101ad6", "generation_seconds_batch": 17.223539856029674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.559302568435669, "target_z": -1.1269531719447203, "ce": 6.197352409362793, "genuineness": 4.68307638168335, "identity": -0.05071478709578514, "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/casual_not_formal/66788c625004026c856e345211c2e71b4624ca939eba9f6db5c39bf0e742fabf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "66a32a5bdeddf3087edb36201898b8ece7361a311c064d25f5d40604df731956", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40a1d495ff4fb9b1acd9a839dba61747ba6dd2257af54dec6ded99929e1cdcba", "generation_seconds_batch": 23.217389811063185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718790054321289, "target_z": -0.8652343069111841, "ce": 6.197352409362793, "genuineness": 4.645420074462891, "identity": 0.5351979732513428, "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/casual_not_formal/66a32a5bdeddf3087edb36201898b8ece7361a311c064d25f5d40604df731956.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "67776f68484852fa3632fd15dd211355bcad45e9a973297166f081bad600ae7e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ee198f365706eef5021bcaec69371c415a8b608dde3824ff20bf4129a49c7d0", "generation_seconds_batch": 16.216135946917348, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.324343204498291, "target_z": -0.5166015704852622, "ce": 6.221180438995361, "genuineness": 4.6956281661987305, "identity": 0.6189442276954651, "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/casual_not_formal/67776f68484852fa3632fd15dd211355bcad45e9a973297166f081bad600ae7e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "6777702ae39a1cb233a103f058eaebd4fe279535cf28ea4676d0f2b251b039a2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66e0d1f2e9b001a90ca24e929b9e7813d92e33cf39adc44705228bd157f08333", "generation_seconds_batch": 21.247167141991667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.619110345840454, "target_z": -0.9121093236388544, "ce": 6.125868320465088, "genuineness": 4.099404335021973, "identity": 0.8719526529312134, "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/casual_not_formal/6777702ae39a1cb233a103f058eaebd4fe279535cf28ea4676d0f2b251b039a2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "68531fcd710ffbd60c2fa66dad9f541edaa844fa57b52e07b221d71d88a74db9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5399263f945a2bb06e9d47043e38f2e72f396b578bba12a17dec90536945d2ae", "generation_seconds_batch": 24.937223019078374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2670276165008545, "target_z": -0.5292968608895305, "ce": 5.959072113037109, "genuineness": 4.557555198669434, "identity": 0.7771462798118591, "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/casual_not_formal/68531fcd710ffbd60c2fa66dad9f541edaa844fa57b52e07b221d71d88a74db9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "685cdec6535cb3381a7213420ad29998f21132b0880b06b6966af65b5e0202d5", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95006e93e3891d42766304637026d62c81bc9ce38bac8e47b6942b7b606a0b5c", "generation_seconds_batch": 17.648491839063354, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.358163595199585, "target_z": -1.9843751141707393, "ce": 5.982900142669678, "genuineness": 3.873466968536377, "identity": 0.6398857831954956, "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/casual_not_formal/685cdec6535cb3381a7213420ad29998f21132b0880b06b6966af65b5e0202d5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "6907a2d02e369cd15083da5f4905ec2bbe73ff6266025dc5702145cbee28b7ae", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f12ff197777de223e7ceef5be793bab73aec4e096182d390e742022ec41891b6", "generation_seconds_batch": 18.104842783068307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3118834495544434, "target_z": -0.48242186497540324, "ce": 5.990842819213867, "genuineness": 4.425758361816406, "identity": 0.7630385756492615, "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/casual_not_formal/6907a2d02e369cd15083da5f4905ec2bbe73ff6266025dc5702145cbee28b7ae.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "694cb990e3c6a1d6c968f86d06a7e26b01d3a8c0c772f8c02cf99a8260e2d51e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4c9c1bd6cc3c31047ab3fd7705e785a3703df950bfbcd162fa8e9d984ae8df30", "generation_seconds_batch": 24.74976805096958, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.967988967895508, "target_z": -0.6757811911098013, "ce": 6.268836498260498, "genuineness": 4.250029563903809, "identity": 0.1847512423992157, "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/casual_not_formal/694cb990e3c6a1d6c968f86d06a7e26b01d3a8c0c772f8c02cf99a8260e2d51e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "69b02a2bc33648cd92b7116cdcabe4c023915425713520a7f2cad7fb0df3dda5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7804f21265a4c313a1a81d773688d15144bcd99f05aabfe515b25a5a7b1e86ec", "generation_seconds_batch": 27.12334181694314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.496290683746338, "target_z": -0.21923822035188145, "ce": 5.661221981048584, "genuineness": 4.306513786315918, "identity": 0.65483158826828, "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/casual_not_formal/69b02a2bc33648cd92b7116cdcabe4c023915425713520a7f2cad7fb0df3dda5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "6a3100aea3dd3e5285c31712e1b8f1d3028b8a0104c88d5ef2dfdb0ec7b432fe", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "719b60dd5b7b428cabc8eb19877f5809994bde0a54624434b21ee76eb2f1fb03", "generation_seconds_batch": 17.694343962939456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524414539337158, "target_z": -0.9335938507693974, "ce": 5.9431867599487305, "genuineness": 4.331617832183838, "identity": 0.5740883946418762, "hypothesis": "La puerta sigue me harta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "wer": 0.13333333333333333, "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/casual_not_formal/6a3100aea3dd3e5285c31712e1b8f1d3028b8a0104c88d5ef2dfdb0ec7b432fe.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "e8cfabadf8bccfb61daa83a220b5698fb766a4d5cc3b57037391016915cd07ae", "generation_seconds_batch": 17.573292694054544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.8710937719204496, "ce": 6.395919322967529, "genuineness": 4.068024158477783, "identity": 0.32029396295547485, "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/casual_not_formal/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "6aa37044ce9b24a79fd76fcef14375e2b59750b2f33bf3ecd0d0f8e38093e3ea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d85114e83ee48dd5218f8b9d59d8cbd51f58a2c2654c8913a73767a20fa31142", "generation_seconds_batch": 24.005818294011988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.5761719367290989, "ce": 5.748591423034668, "genuineness": 4.3127899169921875, "identity": 0.729060173034668, "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/casual_not_formal/6aa37044ce9b24a79fd76fcef14375e2b59750b2f33bf3ecd0d0f8e38093e3ea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "6ae29dd5c37261400dc1bb670f63ce8bfb2aa6364d25b868ce59fed8c4c33d10", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "146b3bacab5805f1ca66becf61b8365e3ddf1180ede676d53b4c83f2c0c73941", "generation_seconds_batch": 25.62851645902265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.002164363861084, "target_z": 0.1088866913792517, "ce": 6.173524379730225, "genuineness": 3.998987913131714, "identity": 0.27673476934432983, "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/casual_not_formal/6ae29dd5c37261400dc1bb670f63ce8bfb2aa6364d25b868ce59fed8c4c33d10.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "6b05cf58c2b20bb45cba02ccc35c1fcf5a70ca9327eb162cd123e9e07e2e54ee", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "edcff75a67113f8933e908f6211df41774a92740da93b62ff35e2c33d48dbae7", "generation_seconds_batch": 17.60875292599667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.40657901763916, "target_z": -0.4746093630793345, "ce": 6.292664527893066, "genuineness": 4.507347106933594, "identity": 0.5903560519218445, "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/casual_not_formal/6b05cf58c2b20bb45cba02ccc35c1fcf5a70ca9327eb162cd123e9e07e2e54ee.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "6b325e8f1532f1e6ab4b2199d6c77383dd1dcd49a4f2e9e90e1cb9b31dc8ad80", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6afa66d3adfbba104dc788cc6e7aa1556f574697a0d2aa342d1e00bab9bd9452", "generation_seconds_batch": 17.05932228907477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.085824489593506, "target_z": -1.3710938061089228, "ce": 6.268836498260498, "genuineness": 4.281409740447998, "identity": 0.3841525614261627, "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/casual_not_formal/6b325e8f1532f1e6ab4b2199d6c77383dd1dcd49a4f2e9e90e1cb9b31dc8ad80.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "b025f2ceac03d9f6f70838fceb723021af602d063f37d5517e5b78369055fc4a", "generation_seconds_batch": 17.25446208298672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1075401306152344, "target_z": -0.6894531501859473, "ce": 6.229123115539551, "genuineness": 4.482243061065674, "identity": 0.742362380027771, "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/casual_not_formal/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.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": "6bc559a22b971808005bf4187a7e031dd94cf00c795ceb752433abb2b2b0e5bd", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b633cd807c9ef41ae0cae8517e751f08967571edbc6115dd7d52c426135dff99", "generation_seconds_batch": 26.137216016068123, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7536778450012207, "target_z": -0.9062499465803097, "ce": 6.252951145172119, "genuineness": 4.306513786315918, "identity": 0.4954555034637451, "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/casual_not_formal/6bc559a22b971808005bf4187a7e031dd94cf00c795ceb752433abb2b2b0e5bd.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "6bebc214288526b70e31fe5d830ef497d67469a903b4d22badd1c07bfcb1d79b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1447bdc0c0fbf54998d06c37337b2ccfa3ddafcedbe552160dabd283be445843", "generation_seconds_batch": 17.510748381959274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3748953342437744, "target_z": -1.1093749847898275, "ce": 6.300607204437256, "genuineness": 4.356721878051758, "identity": 0.6315491199493408, "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/casual_not_formal/6bebc214288526b70e31fe5d830ef497d67469a903b4d22badd1c07bfcb1d79b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "6c538abcc6a7020ae34006d470b50545440bb86093191c6d50acf7b20e24889a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46f5f1e92a211be2914630e52b7a2f8c65260c0870de6be4eb150e1e57eeb65d", "generation_seconds_batch": 23.2453904079739, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.438262462615967, "target_z": 0.39062494383184965, "ce": 6.252951145172119, "genuineness": 4.419482707977295, "identity": 0.3880882263183594, "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/casual_not_formal/6c538abcc6a7020ae34006d470b50545440bb86093191c6d50acf7b20e24889a.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": "6c7fe8008c85ea4bd3c763de7bea72a5f4ed8b8ba6aeba73e50467770f8bc397", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bfb2aa3ba45354d1ed0f6e6e6406c0b860aed26ef9f52d6bdc5d049641369e3b", "generation_seconds_batch": 17.67508417600766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.533670425415039, "target_z": -0.2324218905843441, "ce": 5.959072113037109, "genuineness": 4.256305694580078, "identity": 0.7665253281593323, "hypothesis": "Las llaves están en la mesa de la cocina, miren el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/6c7fe8008c85ea4bd3c763de7bea72a5f4ed8b8ba6aeba73e50467770f8bc397.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "6d013d2d7b1a99ac845debd65e367689f8b21c304b304b5991191a89cbb55126", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fde2ee96e1f460257d941ee8a65ad62fb5ed2ade002ab10e442ce6011dedd8d", "generation_seconds_batch": 18.12268712592777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8166897296905518, "target_z": -1.5859375000448204, "ce": 6.133810997009277, "genuineness": 4.293961524963379, "identity": 0.439867228269577, "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/casual_not_formal/6d013d2d7b1a99ac845debd65e367689f8b21c304b304b5991191a89cbb55126.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "6d04cbae42444cd1a966c10fd178052b30881e6e323c7a221f77f8e0f94688d8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98aa2d816d2a336921c0766bbb2e1e8d088ae4956713009ce09c09b223ed602d", "generation_seconds_batch": 21.965033459011465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3748953342437744, "target_z": -1.265624992178815, "ce": 6.252951145172119, "genuineness": 4.306513786315918, "identity": 0.6131911277770996, "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/casual_not_formal/6d04cbae42444cd1a966c10fd178052b30881e6e323c7a221f77f8e0f94688d8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "6e1ed2e7c377d65c9502022de4454fbe909677d28f9cf3d4372a68886bbd8729", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "974b30b19c39cf3f2a5e0e0529dbb345787c1269d00cb2c3b2140964ff4a730e", "generation_seconds_batch": 23.158721837913617, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192267894744873, "target_z": -0.4238281407168046, "ce": 6.1020402908325195, "genuineness": 4.394378662109375, "identity": 0.7804980278015137, "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/casual_not_formal/6e1ed2e7c377d65c9502022de4454fbe909677d28f9cf3d4372a68886bbd8729.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": "6e2b127d050a2f49bc76771a9d69b537789611523ba570dbdaeefb5c038b9f42", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "077e8089c71ab2b60be655ce13bbd22412202247fa1b31b8476e346da669afa5", "generation_seconds_batch": 16.483289979980327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1449201107025146, "target_z": -0.6904297062429916, "ce": 6.427689552307129, "genuineness": 4.432034492492676, "identity": 0.4920938014984131, "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/casual_not_formal/6e2b127d050a2f49bc76771a9d69b537789611523ba570dbdaeefb5c038b9f42.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": "6e336a54df8e8f0cc5f8493c83f2e57cfbff7840944468fe6def2d3b4e900a30", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "fa3c2f287fc1d7a6bf5c0563bbf200753481b21ecaa2950383527041ce70cb8a", "generation_seconds_batch": 17.623624671949074, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.323631286621094, "target_z": 0.3408203806476437, "ce": 6.205295085906982, "genuineness": 4.482243061065674, "identity": 0.4444897174835205, "hypothesis": "Und deine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sagt mir, wann wir reden können.", "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/casual_not_formal/6e336a54df8e8f0cc5f8493c83f2e57cfbff7840944468fe6def2d3b4e900a30.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "6e81c7003c2d03db6f8ad04f63901d944c0d4b17171911746175bb235715c10f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a06d3e3bc00d35a559b5c427871fa6fc04c985a676c8a0397bf2b2dafc65e4ba", "generation_seconds_batch": 23.580471798893996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6950266361236572, "target_z": -0.05981446855847654, "ce": 5.990842819213867, "genuineness": 4.362998008728027, "identity": 0.48025771975517273, "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/casual_not_formal/6e81c7003c2d03db6f8ad04f63901d944c0d4b17171911746175bb235715c10f.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": "6eaca867f9ac2a999e3886c8530db73c001ce7e2510663d52709a8ec313b98ee", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcb5187e8431cc8923e645af93d4db02d156660caea6ea7960165c5a6255687e", "generation_seconds_batch": 23.223817850928754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.504478931427002, "target_z": -1.1757812414136617, "ce": 5.792276382446289, "genuineness": 4.118232727050781, "identity": 0.09849635511636734, "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/casual_not_formal/6eaca867f9ac2a999e3886c8530db73c001ce7e2510663d52709a8ec313b98ee.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "6f0f342a399f4d9c0213bfa1b1e415bbbeab4d9d78a5e370e7b6aec270713663", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "346d9943b4663bcf5cc3c9f62921d72f3ab934d1d4a5576337d3aeeb5b4f9e20", "generation_seconds_batch": 24.85562670393847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.129255771636963, "target_z": 0.17797844776078786, "ce": 6.070269584655762, "genuineness": 4.482243061065674, "identity": 0.7301598787307739, "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/casual_not_formal/6f0f342a399f4d9c0213bfa1b1e415bbbeab4d9d78a5e370e7b6aec270713663.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "6f9067451ba4411a76a7039e4df7501be546d86a64776677d916a4459a069872", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e985fcdd5ad4c70499f6c17efeec3ec837b25616c7eb6f56f57c19fefca65e97", "generation_seconds_batch": 22.75484538602177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.7304686451407283, "ce": 6.0305562019348145, "genuineness": 4.494794845581055, "identity": 0.6903783082962036, "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/casual_not_formal/6f9067451ba4411a76a7039e4df7501be546d86a64776677d916a4459a069872.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "6fa8bd544756c66e9ff8e4bd4caa7edcc65c1738cc743bdad15f8f9c2f45a17c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "762b95771281fae0e834ad1eca937d9fde47a2b5f4515dde1bb0cf09aec60370", "generation_seconds_batch": 19.162483896012418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9761770963668823, "target_z": -1.4999999695956612, "ce": 5.959072113037109, "genuineness": 4.469690799713135, "identity": 0.5857343673706055, "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/casual_not_formal/6fa8bd544756c66e9ff8e4bd4caa7edcc65c1738cc743bdad15f8f9c2f45a17c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "6ffd0d1237ba47310c30b65b734e6b9838fec07cfd08a8e21c4a71c68f6ff866", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "89b2193877536fe9d50d68c44133518dd5394c85a68b247df18191fa2ad37686", "generation_seconds_batch": 24.503014539950527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9530367851257324, "target_z": -0.5546875759740296, "ce": 6.268836498260498, "genuineness": 4.281409740447998, "identity": 0.31737375259399414, "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/casual_not_formal/6ffd0d1237ba47310c30b65b734e6b9838fec07cfd08a8e21c4a71c68f6ff866.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "701e8799388abdb270fa669408bf15f9ca3eadf212e711087507e44d4453a1eb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7defe7c3aa80b8041a830e29b618e6aa2bb98b040522edbed95300cef4c731d", "generation_seconds_batch": 24.95610836800188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.873293161392212, "target_z": -0.7773437790485174, "ce": 5.824047088623047, "genuineness": 4.319066047668457, "identity": 0.6446892023086548, "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/casual_not_formal/701e8799388abdb270fa669408bf15f9ca3eadf212e711087507e44d4453a1eb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "70c1369abefa1cd2459481d2d380a45cb24d020c5f552249a7d68a0438b88d60", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb61be1b3ce8b7e9181a9b8ac417a0312615ff9e8f7708f6f4917443d6cb9915", "generation_seconds_batch": 18.310392709099688, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3150875568389893, "target_z": -1.0019531780779327, "ce": 6.173524379730225, "genuineness": 4.231201171875, "identity": 0.7423811554908752, "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/casual_not_formal/70c1369abefa1cd2459481d2d380a45cb24d020c5f552249a7d68a0438b88d60.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "70dd11795354e7cdb05d9f7b8df2fc86e5034f62c7dc02e45dbcbd392bf9271a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afc9113db437a84ed09ca0e679049562d9afb1b36f06f7f9dc9df161740d1a73", "generation_seconds_batch": 23.97235517203808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.0351562059782202, "ce": 6.229123115539551, "genuineness": 3.9550557136535645, "identity": 0.3084127902984619, "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/casual_not_formal/70dd11795354e7cdb05d9f7b8df2fc86e5034f62c7dc02e45dbcbd392bf9271a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "71630feb3f0e37e4a242b5a761dbc98b347324b3ceeeb135b82b91e48dc7a896", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e30db03248380211407c6b4f736e853cd606bb75c1c021a66dce15152c0cba06", "generation_seconds_batch": 18.95441730809398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.677138328552246, "target_z": -1.828124983077042, "ce": 6.427689552307129, "genuineness": 4.1998209953308105, "identity": 0.5235334634780884, "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/casual_not_formal/71630feb3f0e37e4a242b5a761dbc98b347324b3ceeeb135b82b91e48dc7a896.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": "71698b11ddc82194e797aa309a094c3694358e5d4a60e4b9974dbfb9b1aa0c4e", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "854352820e9ee3d45af564437e16cbef31ec07c90b38754597eae567ade0111e", "generation_seconds_batch": 20.70403451297898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3991029262542725, "target_z": -0.4658203423808991, "ce": 6.038498878479004, "genuineness": 4.595211505889893, "identity": 0.5878300666809082, "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/casual_not_formal/71698b11ddc82194e797aa309a094c3694358e5d4a60e4b9974dbfb9b1aa0c4e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "71ae9e281d5b9d73f1de7d25f2c7ad433ddbf3344a15c19a682a060cfdd4824e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "720c94d47d3b4751e513f270356c21131c245b5797cbfa84f6c0f786aa8ab015", "generation_seconds_batch": 17.588289493927732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1025562286376953, "target_z": -0.5937500200139751, "ce": 6.054384231567383, "genuineness": 4.632867813110352, "identity": 0.8016917109489441, "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/casual_not_formal/71ae9e281d5b9d73f1de7d25f2c7ad433ddbf3344a15c19a682a060cfdd4824e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "71f0da7091ddb8276ebe0d8bea54e3d4b78d0d1c40af57f79448fdbcf37c7b78", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f49579ada72838a74d0d86603559a07d7baffc47db35b6c793872e780be4355e", "generation_seconds_batch": 18.01514836505521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.758572578430176, "target_z": -0.19116204893929328, "ce": 6.038498878479004, "genuineness": 4.607763767242432, "identity": 0.7240591645240784, "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/casual_not_formal/71f0da7091ddb8276ebe0d8bea54e3d4b78d0d1c40af57f79448fdbcf37c7b78.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "71fceaa179f66c45dd4d3cc8eabf1605e9c6c28cefa0200ee933db01df77845f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b06dbcbf835b609e6d3c85251a7b73cb05fe9b1818af40fed5f041b46a33e6f7", "generation_seconds_batch": 17.073029723018408, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8358242511749268, "target_z": -0.09545895142269224, "ce": 6.260893821716309, "genuineness": 4.65797233581543, "identity": 0.6396275758743286, "hypothesis": "Ouvirent la petite boîte. Elle était près de lavier et fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.21052631578947367, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/71fceaa179f66c45dd4d3cc8eabf1605e9c6c28cefa0200ee933db01df77845f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "72038fabfc331198f7ac7c9d1d6d9274e59ffdf40f677c3d9024321f9c61f6a7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "7464d855ade9c102b1882eba952123b026763080401ad30c19dd7a90eca9028f", "generation_seconds_batch": 26.26931932102889, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4875686168670654, "target_z": -0.30126955692617047, "ce": 5.959072113037109, "genuineness": 4.494794845581055, "identity": 0.5955042839050293, "hypothesis": "J'ai trouvé la tetite boite. Elle était près de la vieille fenêtre.", "wer": 0.47368421052631576, "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/casual_not_formal/72038fabfc331198f7ac7c9d1d6d9274e59ffdf40f677c3d9024321f9c61f6a7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "72298dd7c356a9efa4dec0fe7ed095054382abb2ba2043675d7086b69f55808a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a475707d929b9aa8a3c661f2f90b82fe5d511a3636055aaa3bb91ef5764825eb", "generation_seconds_batch": 16.491749526001513, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9961129426956177, "target_z": -1.2773437659993347, "ce": 6.173524379730225, "genuineness": 4.017816066741943, "identity": 0.3355504870414734, "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/casual_not_formal/72298dd7c356a9efa4dec0fe7ed095054382abb2ba2043675d7086b69f55808a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "72aa594ff9080fad71f6117a27bbbec028ad23b1a16dbbbdf4e3b6b3df285bfa", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c6a5ec17f01a3806cbcd0a8586ea53e853f870239a28755876fbb98c5a911186", "generation_seconds_batch": 17.678491902071983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.583510160446167, "target_z": 0.04443359834830722, "ce": 6.038498878479004, "genuineness": 4.281409740447998, "identity": 0.13689956068992615, "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/casual_not_formal/72aa594ff9080fad71f6117a27bbbec028ad23b1a16dbbbdf4e3b6b3df285bfa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "73176114adf46154bfe402d5710814c42bb532175973eb85dbba8e9d5150564f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c857c034bf7fc2abe57db6d81d4b2da1fb72041a42d09584ef2b0d9556688804", "generation_seconds_batch": 23.440093860030174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.147412061691284, "target_z": -0.41601564586716594, "ce": 5.824047088623047, "genuineness": 4.425758361816406, "identity": 0.6428133249282837, "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/casual_not_formal/73176114adf46154bfe402d5710814c42bb532175973eb85dbba8e9d5150564f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "732a189a137fbbac5f3dde3b8184109daf226a422c545bbf66ff217b077f673d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c20713e910c5e7c1f7770b58fdaeefb6dc8be8f3dcb68b1e35fab3b3c2d34252", "generation_seconds_batch": 23.575432594981976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0260167121887207, "target_z": -1.5273438129760712, "ce": 6.38003396987915, "genuineness": 4.293961524963379, "identity": 0.6052496433258057, "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/casual_not_formal/732a189a137fbbac5f3dde3b8184109daf226a422c545bbf66ff217b077f673d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "733cdee3e94baa19b2c06cf27bcecd6863cd9020c6b42183d021367634d10061", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d519d228ddda75774a613295305a37d155d9740c5fcae64d322319039ac6af34", "generation_seconds_batch": 23.97235517203808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.51444673538208, "target_z": -0.9726562508339827, "ce": 6.221180438995361, "genuineness": 4.118232727050781, "identity": 0.41036009788513184, "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/casual_not_formal/733cdee3e94baa19b2c06cf27bcecd6863cd9020c6b42183d021367634d10061.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "739e46171a5ceafa3b304483fb1f811a7c0fa378f69510b15bf91a1c58dc9789", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8271f9e3d58efa2c160da6f7c32d4957da0785048c9ce5714d8e1512d0ecd283", "generation_seconds_batch": 16.4145659430651, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.078125015046897, "ce": 6.213237762451172, "genuineness": 4.005263805389404, "identity": 0.3854225277900696, "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/casual_not_formal/739e46171a5ceafa3b304483fb1f811a7c0fa378f69510b15bf91a1c58dc9789.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "741410c942b355c69129ef17d7c4e0e1db40481efa94b6ebf8486970a3568fb0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a75f008dac26940a331e718e0087d73eab3c1eadcf4e6e7b03505d113f1e41a", "generation_seconds_batch": 24.41254854190629, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1847920417785645, "target_z": -0.3837889896417982, "ce": 6.117925643920898, "genuineness": 4.406930446624756, "identity": 0.6145573854446411, "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/casual_not_formal/741410c942b355c69129ef17d7c4e0e1db40481efa94b6ebf8486970a3568fb0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "741afc7b012dd98666b0e079daddad42bdefbe52770b67d1ca7f8839b32b7e13", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53fc13b6daf92389592263e94c6ecb5c707173971542b7596b6c7de970950dbc", "generation_seconds_batch": 15.289207939989865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.833421468734741, "target_z": -0.9531249648734962, "ce": 6.173524379730225, "genuineness": 4.369274139404297, "identity": 0.6940515637397766, "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/casual_not_formal/741afc7b012dd98666b0e079daddad42bdefbe52770b67d1ca7f8839b32b7e13.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": "74da83c359c2da94aaa5c74dbbc7751cdd5e5c8b5c08b3e6b1acf1e441eed556", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4818bff2e2c67f703dc4766540323ede615c3500d7c3700141cc9e71adb937b", "generation_seconds_batch": 24.893112823017873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7586617469787598, "target_z": -1.0019531358641798, "ce": 5.919358730316162, "genuineness": 4.507347106933594, "identity": 0.7634755373001099, "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/casual_not_formal/74da83c359c2da94aaa5c74dbbc7751cdd5e5c8b5c08b3e6b1acf1e441eed556.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "751adf85d7afd377222f37761c38543a54befc9df46bf9cfa52a697a394113bd", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9faa044dbae800e49e9522f19ffac6a029e46938ca1c198c3b8bf7965a358bf3", "generation_seconds_batch": 23.134691886953078, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.137444019317627, "target_z": -0.5566406559886071, "ce": 5.9431867599487305, "genuineness": 4.168440818786621, "identity": 0.7103195786476135, "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/casual_not_formal/751adf85d7afd377222f37761c38543a54befc9df46bf9cfa52a697a394113bd.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": "752294423c10af915e38aa83518e74eae3f4b2914abd84a8eff25fbe0c2f66fc", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5076547cb5b1f539f327c453d573b555f81a81c9559f76b5097f59d3e26095fc", "generation_seconds_batch": 24.029069230891764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5486223697662354, "target_z": -0.2753905954572568, "ce": 6.117925643920898, "genuineness": 4.2374773025512695, "identity": 0.5618623495101929, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunchstruk.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/752294423c10af915e38aa83518e74eae3f4b2914abd84a8eff25fbe0c2f66fc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "7586d92af73e64f0b0eecf8e41bc99570c8beeeb49a436e5e2766dbca7d10466", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92f2bc706eb0c2c4892db4723df4115515c139173ee04f33aaf1813c2186461f", "generation_seconds_batch": 23.707004179013893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.1766037940979, "target_z": 0.2812500308449416, "ce": 6.260893821716309, "genuineness": 4.400654315948486, "identity": 0.3436373770236969, "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/casual_not_formal/7586d92af73e64f0b0eecf8e41bc99570c8beeeb49a436e5e2766dbca7d10466.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "75a31980e050998d33575d9a42c90a94b0a0712efa4a73f2a2d531ec264488d4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "2111f8d477fd3722bdc05d646e641a83e0feb4277594d2ffbb38afe201ee1234", "generation_seconds_batch": 22.728337771026418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.7343749364428624, "ce": 6.205295085906982, "genuineness": 4.394378662109375, "identity": 0.6213375926017761, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/75a31980e050998d33575d9a42c90a94b0a0712efa4a73f2a2d531ec264488d4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "75b4c166671154044cb0acc26f9d7041ac828c9f24616b9478ef839d0304f4d8", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3e122afb43a1e53299fffb8d6e47bce6c5d4a752cedcab2998e1d388b7b86ff0", "generation_seconds_batch": 19.03715557209216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0559206008911133, "target_z": -1.5078125918693164, "ce": 5.796247482299805, "genuineness": 3.9676074981689453, "identity": 0.6946821212768555, "hypothesis": "J'ai trouvé la petite boite. Elle était prêt de la vieille fenêtre. Se procevance to fit as titres.", "wer": 0.47368421052631576, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/75b4c166671154044cb0acc26f9d7041ac828c9f24616b9478ef839d0304f4d8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "75ebb2270bb17937ee849370e4dffe811f522fc29b9deb2a9bfad24dfef40b1e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "549ed7c4bc4ae81ebb511788afbdbabd665250b3e84a3033d3b982de751d409c", "generation_seconds_batch": 24.258394913980737, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.800001859664917, "target_z": 0.04260258827356045, "ce": 6.260893821716309, "genuineness": 4.507347106933594, "identity": 0.6197860240936279, "hypothesis": "Ik heb het klein, kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/75ebb2270bb17937ee849370e4dffe811f522fc29b9deb2a9bfad24dfef40b1e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "75edb4a1869868f9ee3cc787cd58c58aafc4c00b0acd5c76d2bb07802ab5bf95", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e9be25dfbc8a2956f7026ffaa9178b5753918dd4a236199d36a256849584a61", "generation_seconds_batch": 25.412665425101295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.044528007507324, "target_z": 0.1425780256728627, "ce": 6.062326908111572, "genuineness": 4.68307638168335, "identity": 0.5927131175994873, "hypothesis": "Wouh! J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/75edb4a1869868f9ee3cc787cd58c58aafc4c00b0acd5c76d2bb07802ab5bf95.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "7635281df13fbedeea9bbc46cfcdd13a8441d2a641560893f30fbf5d63d17075", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57aaf5c8c8b0fabeffef39851816a32dca63a5da1d0605f6f28de3834555cd62", "generation_seconds_batch": 18.12268712592777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.295151710510254, "target_z": -1.2031250440810073, "ce": 6.332377910614014, "genuineness": 4.344170093536377, "identity": 0.641931414604187, "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/casual_not_formal/7635281df13fbedeea9bbc46cfcdd13a8441d2a641560893f30fbf5d63d17075.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "76566b3f50f8bbf2f3491bd0ff643702956ca989461392d027d69930366c244a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b541faf48f8b0c1016924d3dc24b4847791affca909e49ab78345b8690ffef46", "generation_seconds_batch": 22.76991375302896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.464607000350952, "target_z": -1.2304687235654814, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.6818529367446899, "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/casual_not_formal/76566b3f50f8bbf2f3491bd0ff643702956ca989461392d027d69930366c244a.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": "768f3661ccaf8f24b397f4b0ca521adc1b4d2038cd24e52e3f75373f7694657b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 127, "seconds_generated": 10.16, "mp3_sha256": "f5b5687c9da22c35d3e507b31dc562e69b6d2d7b5e4ed4db1fdd504f63b3f9c0", "generation_seconds_batch": 17.223539856029674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1655683517456055, "target_z": -1.535156177400317, "ce": 5.919358730316162, "genuineness": 4.381826400756836, "identity": 0.4094417095184326, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam neem ze mee.", "wer": 0.13636363636363635, "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/casual_not_formal/768f3661ccaf8f24b397f4b0ca521adc1b4d2038cd24e52e3f75373f7694657b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "76a149589dbe6d0b49970068389e4f112af8ba0f474cf41cb22cbc9197c212f7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c2ca82bf5718f611779dd2e8c79a28a0bd69c13ca2085ff40d7871f8bc2733bc", "generation_seconds_batch": 23.185679672984406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3186469078063965, "target_z": 0.3081053853395602, "ce": 5.855817794799805, "genuineness": 4.6956281661987305, "identity": 0.7439894676208496, "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/casual_not_formal/76a149589dbe6d0b49970068389e4f112af8ba0f474cf41cb22cbc9197c212f7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "76d97a1ca2d6177fe54301c8ae8adbe6f7bf7c43f01711c917c20f109d15b0ea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b0e2040b53329f997fc0c5e7cfc6d424110dd9cf83ed3c52e0ef5092d8b5c303", "generation_seconds_batch": 24.04835745296441, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.414767265319824, "target_z": -1.136718738783657, "ce": 6.221180438995361, "genuineness": 4.344170093536377, "identity": 0.43548619747161865, "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/casual_not_formal/76d97a1ca2d6177fe54301c8ae8adbe6f7bf7c43f01711c917c20f109d15b0ea.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "9c67c2922e03b1058091321b0d19b476a4c5d7f37b223656fcce9ae391a36e6b", "generation_seconds_batch": 16.048109869938344, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3748953342437744, "target_z": -1.324218715393736, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.642220139503479, "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/casual_not_formal/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "77c5b0314b1376b785f6906fdf3523cd8b3937c7604fb236cbf956fdd7431145", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9cf043bc2d87e044ada94fe30a6c3592f03637959c0d3cd0980bf4a4f9fcb4a", "generation_seconds_batch": 25.05234414502047, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7130937576293945, "target_z": -0.23925776938343296, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.5080185532569885, "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/casual_not_formal/77c5b0314b1376b785f6906fdf3523cd8b3937c7604fb236cbf956fdd7431145.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "77c989b34471b1e76843143d0bd51685fc1fc63f1e396f5e286292aec3f227ef", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fdd38872f984d776dcee4133614569441e9577a944544aeeb5fbfe55bbedf24", "generation_seconds_batch": 21.083252437063493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.295151710510254, "target_z": -1.1054687599069413, "ce": 6.24500846862793, "genuineness": 3.9676074981689453, "identity": 0.4376128017902374, "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/casual_not_formal/77c989b34471b1e76843143d0bd51685fc1fc63f1e396f5e286292aec3f227ef.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "78dfbf0428563c1f3d5b6fd4e69eef784572cf173c897abdc58e98440ac0c89a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "237d35b1c443cc0d8f20e4514912baa1f53b951ef5f209ce5c829a2f0b3be334", "generation_seconds_batch": 26.15320115198847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6196441650390625, "target_z": -0.1499023010516245, "ce": 6.141753673553467, "genuineness": 4.356721878051758, "identity": 0.5309135317802429, "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/casual_not_formal/78dfbf0428563c1f3d5b6fd4e69eef784572cf173c897abdc58e98440ac0c89a.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "9142375c00c4cbef4206fd38d047a016f5e9e519fe52fc811ebe6df19ad45913", "generation_seconds_batch": 15.911555426078849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3318192958831787, "target_z": -0.4765625082085627, "ce": 6.1020402908325195, "genuineness": 4.469690799713135, "identity": 0.5658239126205444, "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/casual_not_formal/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.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": "7a4617fbdadc8ee48c2de30882965fccca85fa1d0e93fe3781087129afdabf87", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cfe06cbd2540e9f61d7e928b70a91a89c24c16a6254b74235143bc08880377eb", "generation_seconds_batch": 17.471166706061922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.331107139587402, "target_z": 0.26416022645318427, "ce": 6.157639026641846, "genuineness": 4.419482707977295, "identity": 0.5829653143882751, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/7a4617fbdadc8ee48c2de30882965fccca85fa1d0e93fe3781087129afdabf87.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": "7a98ef70d62aea9625c588dae93cf81a0638ef579014d0eb75ceebed186bc94c", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2781de986cc0e94633e85d20c455195bd716b2c8c146512d55132e0d2df2e36a", "generation_seconds_batch": 24.059360969928093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -0.976562426521514, "ce": 6.117925643920898, "genuineness": 4.7332844734191895, "identity": 0.7728453278541565, "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/casual_not_formal/7a98ef70d62aea9625c588dae93cf81a0638ef579014d0eb75ceebed186bc94c.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": "7a98fefea9465be73f3e892436ef52063470c11f5a2c08c948b311c58bb2ae47", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "20beaf488b844a7baa868e3e8d434883f7aee4ed2efdea842498d6b406156be8", "generation_seconds_batch": 18.08622430008836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.3554687169752584, "ce": 6.23706579208374, "genuineness": 4.13078498840332, "identity": 0.3813970685005188, "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/casual_not_formal/7a98fefea9465be73f3e892436ef52063470c11f5a2c08c948b311c58bb2ae47.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": "7ab41bb9636d5596633a2cbe59f3d60b5b35c911ceee3ec793cff37dc6ea874e", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "437df2afe0d799e60ee61f9a1d030244be57236a85ad5d981114c6c05b3bb32d", "generation_seconds_batch": 18.86433734698221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.671041250228882, "target_z": -0.18151862239874156, "ce": 6.260893821716309, "genuineness": 4.381826400756836, "identity": 0.3104000687599182, "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/casual_not_formal/7ab41bb9636d5596633a2cbe59f3d60b5b35c911ceee3ec793cff37dc6ea874e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "7b28704da582346aef47b82395ccbb09c7768413946afee69ed61a58d54f92e1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1c8952de6ad0707224698ebf417bae7df344163e2d15f4e0fbab8070fea59a6", "generation_seconds_batch": 22.385621860041283, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6871063709259033, "target_z": -1.8085937700603945, "ce": 6.356205940246582, "genuineness": 4.137060642242432, "identity": 0.592456579208374, "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/casual_not_formal/7b28704da582346aef47b82395ccbb09c7768413946afee69ed61a58d54f92e1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "7b7564d3859dd85e14942f164433ef89d44b8670327a5798e44ddcca2fa72ef0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b709dfacbb7ef734754bfc48df234b52e3fccd4c81226fa60ab18e36715b6862", "generation_seconds_batch": 18.01514836505521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.356027126312256, "target_z": 0.30175789742773573, "ce": 6.0305562019348145, "genuineness": 4.7834930419921875, "identity": 0.7022911310195923, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/7b7564d3859dd85e14942f164433ef89d44b8670327a5798e44ddcca2fa72ef0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "7b76ca1fc460372a7f0b1dbe58b14178abed7a20cc71c2fb743e0f796586c265", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9312202fc39ca2f6e1a5f32048da4eb37fe9a90c88db53fe6300cc1f0dbe252d", "generation_seconds_batch": 15.785190475988202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6440300941467285, "target_z": -1.1757813277988662, "ce": 6.173524379730225, "genuineness": 4.086852550506592, "identity": 0.5858582854270935, "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/casual_not_formal/7b76ca1fc460372a7f0b1dbe58b14178abed7a20cc71c2fb743e0f796586c265.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "7b957c8e9b811caca438d596992bc2fb8f4c43de789915be3a7a672ade5bbc2d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a8e4a8af7b460c1680a5df6f843f3413f234b730430123fc625024703792322", "generation_seconds_batch": 14.930282880901359, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5874266624450684, "target_z": -1.746093805260533, "ce": 6.24500846862793, "genuineness": 4.118232727050781, "identity": 0.48180487751960754, "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/casual_not_formal/7b957c8e9b811caca438d596992bc2fb8f4c43de789915be3a7a672ade5bbc2d.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": "7badc9901253e8be7e26283568d06fa49353d7f973d8e4bf13d84b7f355253ce", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6fd606ea1793638f2551cd10070da709cd3c30c19059b14bb3cb3453cf026e0", "generation_seconds_batch": 18.43345167802181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8788108825683594, "target_z": 0.0483398246500431, "ce": 6.117925643920898, "genuineness": 4.519899368286133, "identity": 0.638094425201416, "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/casual_not_formal/7badc9901253e8be7e26283568d06fa49353d7f973d8e4bf13d84b7f355253ce.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "7bb863c07cab6a29e8abf0c9843ed73142405fe47b6f1ef4237c2cf73b43756f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee148e724f61a967d232b2b3aa987a4e2fcb663f2b3fdd584d9a5f7a1d486d60", "generation_seconds_batch": 22.075816667987965, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.751096487045288, "target_z": -0.022460920539925298, "ce": 5.49045467376709, "genuineness": 4.457139015197754, "identity": 0.5620415210723877, "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/casual_not_formal/7bb863c07cab6a29e8abf0c9843ed73142405fe47b6f1ef4237c2cf73b43756f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "7bca4ac4b5b34fe31ca90e39a4c6121b787b7111ff7c71acacf2a6bb120ecbbf", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40e3d3e82889f893ef44ba05739dc1cd2b1d16158cceed8d5c8e6d2d68569234", "generation_seconds_batch": 24.845025430084206, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6988539695739746, "target_z": -1.0800781390368348, "ce": 6.149696350097656, "genuineness": 4.469690799713135, "identity": 0.7478652596473694, "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/casual_not_formal/7bca4ac4b5b34fe31ca90e39a4c6121b787b7111ff7c71acacf2a6bb120ecbbf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "7bd8e157335ae35683a7f8109cec6dc536855e85a72dd2a035ccedabcbb3209a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e97054cd15997ea9287286714515a26753e2eabf29fb8dbf0952620b2c25ee3d", "generation_seconds_batch": 22.55454672104679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.985966444015503, "target_z": 0.15209957805914334, "ce": 6.205295085906982, "genuineness": 4.444586753845215, "identity": 0.42598193883895874, "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/casual_not_formal/7bd8e157335ae35683a7f8109cec6dc536855e85a72dd2a035ccedabcbb3209a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "7c79754c2d3b65a687592c79713c6a9c38026806449f04a1dc2de3da36ec8209", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "94899d5485826dd342df951a21067d785b4ddf40cbee4d2ab7f2a613c756eb81", "generation_seconds_batch": 23.06136308494024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2670276165008545, "target_z": -0.4746093110138665, "ce": 6.149696350097656, "genuineness": 4.231201171875, "identity": 0.42768344283103943, "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/casual_not_formal/7c79754c2d3b65a687592c79713c6a9c38026806449f04a1dc2de3da36ec8209.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "7c852641426ba3f44cc7b5d8116610f487a662ce8062b66aac460ddd7c8a49d5", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "32740d9419263b82cf4dd6e9d73fa325aa231a63352bb7925f3bb7f48f2958fb", "generation_seconds_batch": 15.362856520921923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.833421468734741, "target_z": -0.8554687989232264, "ce": 5.855817794799805, "genuineness": 4.381826400756836, "identity": 0.5814074277877808, "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/casual_not_formal/7c852641426ba3f44cc7b5d8116610f487a662ce8062b66aac460ddd7c8a49d5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "7cb83fc53ee4efb8e768d3e1f5680b5e4baceab9cc94984c901718711852fa05", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9ec57093fadcb4f4964b8a43116eb88c5590d1a075b4622b96930bd19d692b0", "generation_seconds_batch": 22.7830135790864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8133962154388428, "target_z": -0.06909179589575133, "ce": 5.903473377227783, "genuineness": 4.909013748168945, "identity": 0.7942526340484619, "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/casual_not_formal/7cb83fc53ee4efb8e768d3e1f5680b5e4baceab9cc94984c901718711852fa05.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "7d2851b48c46fce88939e19453ac86661c9a49b4b30c9c5c22e05a45fd1b5b7f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8d1865f82136c4fe8eaefa79ff61f338c6d904347df7bc44d09c6b42f429e1c", "generation_seconds_batch": 21.965033459011465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.6562500936429285, "ce": 5.871703147888184, "genuineness": 4.086852550506592, "identity": 0.5561763048171997, "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/casual_not_formal/7d2851b48c46fce88939e19453ac86661c9a49b4b30c9c5c22e05a45fd1b5b7f.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": "7d6da2836c649daeca905c33885d3e8e8ead42f923c0eddd7271400fcfc1d889", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ea9b4c41632098ea5c3be0d2471de651a0b57efc82fdc2e6209f5ecdae42c524", "generation_seconds_batch": 18.350836622994393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.147412061691284, "target_z": -0.7519531144639701, "ce": 6.308549880981445, "genuineness": 4.005263805389404, "identity": 0.2718152403831482, "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/casual_not_formal/7d6da2836c649daeca905c33885d3e8e8ead42f923c0eddd7271400fcfc1d889.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "7e794524b10e16143f8dedcaf085af826b2348a6cfcbcb5173b3529f9bd3b853", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d661b9f1932de47e2f01bf7cb86eee39f9ec77d4518e1b15d2d777459f228a33", "generation_seconds_batch": 22.988013435038738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.513022422790527, "target_z": 0.38281254215292626, "ce": 6.038498878479004, "genuineness": 4.482243061065674, "identity": 0.6258493661880493, "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/casual_not_formal/7e794524b10e16143f8dedcaf085af826b2348a6cfcbcb5173b3529f9bd3b853.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "7ee4b158011a39c0433ea55a506a2e14e32c8b2acc624379df56272f89555ff1", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a6826ce32b1dffed751fafc9ae843dd7ea68b0a57a8515db3257b93c00fbd2b", "generation_seconds_batch": 17.82586977595929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6290783882141113, "target_z": -1.0136717895032046, "ce": 6.165581703186035, "genuineness": 4.482243061065674, "identity": 0.7521325349807739, "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/casual_not_formal/7ee4b158011a39c0433ea55a506a2e14e32c8b2acc624379df56272f89555ff1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "7f45b2c812b87316afd81f7006f9b608f6c1831736dd2a061496523116392b23", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e51cc33d15c744c84f8e22fb70b5e96b29a0845b3df644f504a7bf67a7c1fdcb", "generation_seconds_batch": 21.516050491016358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.614037036895752, "target_z": -0.2077637046672333, "ce": 6.292664527893066, "genuineness": 4.7834930419921875, "identity": 0.5953596234321594, "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/casual_not_formal/7f45b2c812b87316afd81f7006f9b608f6c1831736dd2a061496523116392b23.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "7f4c3a11ad4f5b5ad19c0833d2a544dda81e870cefc52d533d263d0797ccdb30", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eb3d14d73e22de4ac3bb0df2b1ecdd88703a97ea54effb530e9538de5a40e73c", "generation_seconds_batch": 17.241436044918373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.364927291870117, "target_z": -1.324218766154607, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.7064526677131653, "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/casual_not_formal/7f4c3a11ad4f5b5ad19c0833d2a544dda81e870cefc52d533d263d0797ccdb30.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "7f6a64696c9156bc885d93ab9be373a79175137d921813d82b99fe2a7eb78598", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d63b38ddb15e387b703aa2fc19007c6fccca9b32c562626af85f4a398d1131b", "generation_seconds_batch": 27.279324246919714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6938700675964355, "target_z": -1.0078124626836922, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.5902652144432068, "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/casual_not_formal/7f6a64696c9156bc885d93ab9be373a79175137d921813d82b99fe2a7eb78598.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "2e96465e6147fd374ffdbc4260c3f8dd72875469f532ac3ae069a147ba0fde13", "generation_seconds_batch": 16.493021990987472, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0078606605529785, "target_z": -0.8066406122741661, "ce": 6.467403411865234, "genuineness": 4.337893962860107, "identity": 0.283308207988739, "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/casual_not_formal/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "7f9b7667ef2d2cdcfda3b78111efddb7c50e1c3e913f3d7289796b0fba24e8cc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c29913c1f19cc2ff1f2d277521087c3bcbba1cf56e87fa8751915638586c2ab", "generation_seconds_batch": 22.106512819998898, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.111812114715576, "target_z": 0.24755863525322824, "ce": 6.117925643920898, "genuineness": 4.20609712600708, "identity": 0.35286152362823486, "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/casual_not_formal/7f9b7667ef2d2cdcfda3b78111efddb7c50e1c3e913f3d7289796b0fba24e8cc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "803124a11ce3d4d2e295579ebe973e28d4c423ff974a7a2b6457376113a51277", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6bd7dc1e3aae09ea999e5d7ffaf6d37e55a8182d083c6cb6d605cd4831e1b47", "generation_seconds_batch": 18.77852300100494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.828437328338623, "target_z": -0.5527344373693928, "ce": 6.268836498260498, "genuineness": 4.231201171875, "identity": 0.25775331258773804, "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/casual_not_formal/803124a11ce3d4d2e295579ebe973e28d4c423ff974a7a2b6457376113a51277.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "80615244929826d94c301d95249e468a6f8f465a7e707fbe9e3713fe5784e84e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8b5fdddaba2c021559ee920f4adf585e68f4f790e87f672a4aa2c5e575496dd9", "generation_seconds_batch": 26.45920089504216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.7167968885862281, "ce": 5.95112943649292, "genuineness": 4.2751336097717285, "identity": 0.5435526371002197, "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/casual_not_formal/80615244929826d94c301d95249e468a6f8f465a7e707fbe9e3713fe5784e84e.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "67d0e19a6b39a1d84caa32450ee9c870aaf15118cc7036a9fee1403c2370b921", "generation_seconds_batch": 15.598526270012371, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.929896831512451, "target_z": -0.025390571609923446, "ce": 6.2767791748046875, "genuineness": 4.6705241203308105, "identity": 0.6779032349586487, "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/casual_not_formal/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "814d8c6cde81b408c45664ffc430b36ece79609034c9a4d7ff88165a8321d157", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f012a8f6505342e4347661fa9752f97ff2d7cac594f8e048288dcf71a81223b", "generation_seconds_batch": 23.331219194922596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -1.1914062732618638, "ce": 6.38003396987915, "genuineness": 4.2751336097717285, "identity": 0.5683609247207642, "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/casual_not_formal/814d8c6cde81b408c45664ffc430b36ece79609034c9a4d7ff88165a8321d157.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "81bc0e2ead7bc2f17ddffb48e59ae61824fcd94b893b60b1be4e53bdb736f2ca", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4bc10a8337bb7a780320045ef935871541d841faf138e13bf81e5c5e2edbf6d", "generation_seconds_batch": 17.03976545901969, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2894554138183594, "target_z": -0.40722656736142926, "ce": 6.141753673553467, "genuineness": 4.231201171875, "identity": 0.28761231899261475, "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/casual_not_formal/81bc0e2ead7bc2f17ddffb48e59ae61824fcd94b893b60b1be4e53bdb736f2ca.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "8210682762be3db12693a3c9732e5fea409037c3782ed5319fb3b7e82ada88f3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "60052dc46d66873739034f8e86311ad70db7a8b318b927ae266d343ac177ec7c", "generation_seconds_batch": 18.420794353005476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1499040126800537, "target_z": -0.49316404300466354, "ce": 6.205295085906982, "genuineness": 4.444586753845215, "identity": 0.5262743830680847, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar, wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/8210682762be3db12693a3c9732e5fea409037c3782ed5319fb3b7e82ada88f3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "822de01939ef1fcf55eca1b4f68eb5763ad61ca0cbcc3dc2b09f008a9405dc6a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a6be83971f3f92ea660ce6771437251f6cf5d463248c55bd9a31cc8088ec3123", "generation_seconds_batch": 22.021132701891474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5842223167419434, "target_z": -0.9765625264777401, "ce": 6.086154937744141, "genuineness": 4.068024158477783, "identity": 0.6086031198501587, "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/casual_not_formal/822de01939ef1fcf55eca1b4f68eb5763ad61ca0cbcc3dc2b09f008a9405dc6a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "8237c0fb4b5e5631ce3c749b1ffa09c861e0ee045858c92346dd918ac8852584", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1391744ea1b995a5d534820f331c1dca0c6f85262cd4d54798eeb7d0b9dd0600", "generation_seconds_batch": 23.217389811063185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0060808658599854, "target_z": -1.3359376033929982, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.6642711162567139, "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/casual_not_formal/8237c0fb4b5e5631ce3c749b1ffa09c861e0ee045858c92346dd918ac8852584.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "82910c0419a28e6abd6963a3dea9a901092675f75e189ad6d63c2a92a95ce67c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05ec895a32702ada46c394ff9140e756f062a89032f1447dd1b8adc1c2c03544", "generation_seconds_batch": 18.817252121050842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.414767265319824, "target_z": -0.9765624989435144, "ce": 6.173524379730225, "genuineness": 4.105680465698242, "identity": 0.40892577171325684, "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/casual_not_formal/82910c0419a28e6abd6963a3dea9a901092675f75e189ad6d63c2a92a95ce67c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "82b7027ad63783119deec6a9a527387c3556c421643d5a5ae3995d3320d3a919", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3d35e1f23adbfefcd1fe2c47ff93895aef25dd167d7306da07154f51d56126f5", "generation_seconds_batch": 22.63391266600229, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.247091770172119, "target_z": -0.48242178400628466, "ce": 6.1020402908325195, "genuineness": 4.482243061065674, "identity": 0.711179256439209, "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/casual_not_formal/82b7027ad63783119deec6a9a527387c3556c421643d5a5ae3995d3320d3a919.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "82c165e48cc5c9981827c641405e31ec4876cba388c6848fe39de412342ada4b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5caee04df63e544e4ad8a18f2be8f7c50797e452af1bda1a3b4fed7fbe6f3ea", "generation_seconds_batch": 14.772496189922094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.2031250289444695, "ce": 6.0305562019348145, "genuineness": 4.005263805389404, "identity": 0.6281550526618958, "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/casual_not_formal/82c165e48cc5c9981827c641405e31ec4876cba388c6848fe39de412342ada4b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "82d2e14497c3d4015f1c4ac2a465caa66e06eb58ae57a3135239eb790c401d87", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2183080ca29ae66e52f5f4f2fbe255ab68849cab976107d55efda59a0c06e161", "generation_seconds_batch": 23.090719264000654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.66387677192688, "target_z": -0.14331056583449286, "ce": 5.708878040313721, "genuineness": 4.557555198669434, "identity": 0.7980561256408691, "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/casual_not_formal/82d2e14497c3d4015f1c4ac2a465caa66e06eb58ae57a3135239eb790c401d87.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "82e66667170534690438264c46ad76fbcc6d987a4c6ef5b5eecd9bc6452c9536", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "48ab4d34210e90e2e72a5122630a30e0e386f0a52d1608c531d19cb2b4a6e513", "generation_seconds_batch": 26.499635227024555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2122037410736084, "target_z": -0.6210937580885086, "ce": 6.038498878479004, "genuineness": 4.570107460021973, "identity": 0.5550099611282349, "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/casual_not_formal/82e66667170534690438264c46ad76fbcc6d987a4c6ef5b5eecd9bc6452c9536.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "839b87ef3c19ed75ec1304e6b9bd7838f61fa6a71b3aac852d3ab3354c52186c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "905a8d869d3ce73f473a3d8e86ba14bcdf19e37f0ff6b9153df48eb20324a739", "generation_seconds_batch": 15.352219147025608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0658886432647705, "target_z": -1.5820311941279874, "ce": 6.435632705688477, "genuineness": 4.444586753845215, "identity": 0.5094915628433228, "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/casual_not_formal/839b87ef3c19ed75ec1304e6b9bd7838f61fa6a71b3aac852d3ab3354c52186c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "8467e40a710d6c593797af3f5ee0270d099c28ce471e3ddfbd4702785b00c433", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3625317424041b85f5b863c95da34488768c350c7c6bbac044a81c509ba4f0a", "generation_seconds_batch": 16.814341040095314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5311784744262695, "target_z": -0.4101562544548415, "ce": 5.839932441711426, "genuineness": 4.3504462242126465, "identity": 0.5502169132232666, "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/casual_not_formal/8467e40a710d6c593797af3f5ee0270d099c28ce471e3ddfbd4702785b00c433.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "84ae096d9d234c8fc35225e99c70b08c3da8a3915b8bba69014462c6619c1337", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ed2f73280b4fdd41bed82f775df127eed18ac6a3218e78da13c3f443912641ab", "generation_seconds_batch": 22.776898788986728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.503054141998291, "target_z": 0.39892571026725765, "ce": 5.990842819213867, "genuineness": 4.432034492492676, "identity": 0.3303586542606354, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/84ae096d9d234c8fc35225e99c70b08c3da8a3915b8bba69014462c6619c1337.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "84b6ec2f2dade9cc873c12b313f479b3126b8d2020c1c42a9e888c288f1360d8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "af6daa1f73e0220075ab62bea890eb0dcafde88e7019d29f814224ea44fd6123", "generation_seconds_batch": 29.448011026950553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.126763820648193, "target_z": 0.13287350880009158, "ce": 6.260893821716309, "genuineness": 4.231201171875, "identity": 0.4891473650932312, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen daar samen.", "wer": 0.16666666666666666, "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/casual_not_formal/84b6ec2f2dade9cc873c12b313f479b3126b8d2020c1c42a9e888c288f1360d8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "851a2c7a3e5312a66db16af052e50368d9ba22b215e26b84b80707f9e3605b3f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73eda6c2b60110f26c4fb74ca5b6e5279425ec71b11c08066538994e4e18250d", "generation_seconds_batch": 22.728337771026418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -1.085937480992885, "ce": 6.109982967376709, "genuineness": 4.494794845581055, "identity": 0.719621479511261, "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/casual_not_formal/851a2c7a3e5312a66db16af052e50368d9ba22b215e26b84b80707f9e3605b3f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "854483de3a1ea84cc428015084fa2b5cfb9cb21f641c74441bc1b4eb084bbfd6", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d0a94a255e35318216ca6c047c6a0b2dea1328232c74bb000f16f981180e228", "generation_seconds_batch": 15.785190475988202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8389391899108887, "target_z": -0.14501952542893048, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.7421693205833435, "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/casual_not_formal/854483de3a1ea84cc428015084fa2b5cfb9cb21f641c74441bc1b4eb084bbfd6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "855283c0d1de067d1966c9c57e1fba0236bec7d7d72ae1f41aedbeeeb489f7a6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab18dcb69865848a03591ba6804c73bc3b5ca71c0d8fea2940cc2d1ee2c63f19", "generation_seconds_batch": 22.578925330075435, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.8085937212572223, "ce": 5.8875885009765625, "genuineness": 4.13078498840332, "identity": 0.25412681698799133, "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/casual_not_formal/855283c0d1de067d1966c9c57e1fba0236bec7d7d72ae1f41aedbeeeb489f7a6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "85576024b0e81a2d72209c52055557242967956ba59a38a776cabd9369a6ebe3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e9a1b948c5fcaabdd96747ecdb3e57eec984c03c55f477a84d61bed029ade53", "generation_seconds_batch": 25.36074345104862, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1997437477111816, "target_z": -0.452148476214549, "ce": 6.268836498260498, "genuineness": 4.469690799713135, "identity": 0.5016670227050781, "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/casual_not_formal/85576024b0e81a2d72209c52055557242967956ba59a38a776cabd9369a6ebe3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "8566a77bfff8088afc54edadec660feacf413a7b9f302a3547624e56a72daa8d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1e72547df638ebbb03d6491ae5e6a646bd7fd8d51d55a1970071822c191aa2d", "generation_seconds_batch": 24.95610836800188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.770409345626831, "target_z": -0.12011720134594832, "ce": 6.1020402908325195, "genuineness": 4.65797233581543, "identity": 0.6676809787750244, "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/casual_not_formal/8566a77bfff8088afc54edadec660feacf413a7b9f302a3547624e56a72daa8d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "8570bea59b0eff7c10b698b1c7f0848e22e2d54291fc1417351add44586d0565", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01b812dd76bb84959dd2256cf6ea1fd21cc73b7e953d2b6d3fcd30af93125a88", "generation_seconds_batch": 17.114031622069888, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7666714191436768, "target_z": -0.060546859682778446, "ce": 6.141753673553467, "genuineness": 4.319066047668457, "identity": 0.8597667813301086, "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/casual_not_formal/8570bea59b0eff7c10b698b1c7f0848e22e2d54291fc1417351add44586d0565.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "857880afff3c2f7e4be7627137f0ad88bc23eddf6c34303492bec7c509c1abdc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c2e45f2f6ac5a4a3da3dad58cf7302c18708f9edbe98f4c161cbd40dbc58f17", "generation_seconds_batch": 22.112723449943587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1773159503936768, "target_z": -0.5664062457284416, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.62501460313797, "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/casual_not_formal/857880afff3c2f7e4be7627137f0ad88bc23eddf6c34303492bec7c509c1abdc.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": "85a8119eaf7b23dd65f9922fb2c497610d0c33fd7830fdc810fab91c13a92dd1", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41e4f05320b654e27de7f227975f2a15afbe6542b81942d4c298ea34d8927965", "generation_seconds_batch": 17.243588243960403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4165468215942383, "target_z": -0.31250004968437783, "ce": 5.974957466125488, "genuineness": 4.381826400756836, "identity": 0.6992759704589844, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine Houtenhek. Ik wacht aan de andere kant.", "wer": 0.14285714285714285, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/85a8119eaf7b23dd65f9922fb2c497610d0c33fd7830fdc810fab91c13a92dd1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "865a0382ac465e93313416586bc3a39fdb8a60c22200a887983657e61acd87bb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4702edbbbd397c01223710a8b9e50bce2b44740cf9781989f8d82bb0917accc0", "generation_seconds_batch": 22.393519739969634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9380850791931152, "target_z": -0.7011718200051873, "ce": 5.982900142669678, "genuineness": 4.093128204345703, "identity": 0.6400601863861084, "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/casual_not_formal/865a0382ac465e93313416586bc3a39fdb8a60c22200a887983657e61acd87bb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "86959055cb8e37d8f928f610c4be18bf80d50515379f84440f3b38b290c5f809", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e0ee3d5d9a2f0bd1d0026e39e96fc0782cda252996587c6d1fc2b99cb23248b", "generation_seconds_batch": 23.575432594981976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7070422172546387, "target_z": -1.863281252734053, "ce": 6.435632705688477, "genuineness": 4.030367851257324, "identity": 0.6367684602737427, "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/casual_not_formal/86959055cb8e37d8f928f610c4be18bf80d50515379f84440f3b38b290c5f809.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": "86efc95245dd2cdd8c5f805da15b0aa727595e471e0c76b804c3e762e051effc", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "cc99e9a9d53b7bd79346139d2d7ebd8ee7593cb83c7831dcffe50b714f4a4adc", "generation_seconds_batch": 17.02780994400382, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5361623764038086, "target_z": -0.2656250248665999, "ce": 6.086154937744141, "genuineness": 4.419482707977295, "identity": 0.6221339702606201, "hypothesis": "Die Straße führ hinter der Brücke nach links. Geh weiter bis zu dem kleinen Holztor. Ich warte auf der anderen Seite.", "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/casual_not_formal/86efc95245dd2cdd8c5f805da15b0aa727595e471e0c76b804c3e762e051effc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "8705ffa5c06761b62e64706642c394f42ed57043b7365626c0a042f26b471adb", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "190352692f1d11dc75d1de97fe8222d7703851a79af70cdf77a92dfa6e70eb13", "generation_seconds_batch": 16.736226157983765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.167968699914641, "ce": 5.760505676269531, "genuineness": 3.9425034523010254, "identity": 0.5270605087280273, "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/casual_not_formal/8705ffa5c06761b62e64706642c394f42ed57043b7365626c0a042f26b471adb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "871044bcec47d63e68d896477c2b8b728141e7ae71827a5434bf52ecff0b5b03", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83c595b1ed66ed7245941b4c6dff5446e855d20408af800bf7ec7a6e0dfb0223", "generation_seconds_batch": 16.392561442917213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4347031116485596, "target_z": -0.9960937713329792, "ce": 6.125868320465088, "genuineness": 4.18099308013916, "identity": 0.35488951206207275, "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/casual_not_formal/871044bcec47d63e68d896477c2b8b728141e7ae71827a5434bf52ecff0b5b03.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "89154830ccb07feacdbdef8a8d735821e61196567bb883f3d401abbec2324924", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d82d2963006ca21ff98489d74f9c4b40c412dc8eb0f36f9e9858188219294165", "generation_seconds_batch": 20.699485231074505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.194047451019287, "target_z": 0.26220696952590195, "ce": 6.149696350097656, "genuineness": 4.532451152801514, "identity": 0.3838599920272827, "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/casual_not_formal/89154830ccb07feacdbdef8a8d735821e61196567bb883f3d401abbec2324924.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "897ae94b8bf8dd2d0bdfd5ac27f1b54fab21f8684439aef53de1ec09c3151891", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2ee5813833744c45b2603bee5aecae0c3c35a8beb2177e0afeb8653b5123974", "generation_seconds_batch": 22.544980152044445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5973947048187256, "target_z": -1.7617188025280792, "ce": 6.348263263702393, "genuineness": 4.344170093536377, "identity": 0.5334153175354004, "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/casual_not_formal/897ae94b8bf8dd2d0bdfd5ac27f1b54fab21f8684439aef53de1ec09c3151891.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "89bc97b55b20efdf25bf407330c0bdaf0aab6f6b25cbceee2d22b7f4dd66d919", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10d3bfd76c0d7654664e5ad5e04fcc5cc60913eaa3f1ca222e9dfdf388f0a97a", "generation_seconds_batch": 16.046525139012374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -1.0546874281286527, "ce": 5.831989288330078, "genuineness": 4.15588903427124, "identity": 0.75186687707901, "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/casual_not_formal/89bc97b55b20efdf25bf407330c0bdaf0aab6f6b25cbceee2d22b7f4dd66d919.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "89c5a820d2f709890faf009d24df1f87ae346dfd075a02ef21d46e44d777f6b8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "611093d658c1c34706558d5d87a96de3a895a5e585174b8a0da8f2d2fb501e01", "generation_seconds_batch": 21.46840405999683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624094247817993, "target_z": -1.0507811881092286, "ce": 6.356205940246582, "genuineness": 4.193545341491699, "identity": 0.33400529623031616, "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/casual_not_formal/89c5a820d2f709890faf009d24df1f87ae346dfd075a02ef21d46e44d777f6b8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "89e87f3b3dfc9454da2c39be6785e17aac00fae3bc138c40f1ce927df2b333cc", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21cb4b92302486b1ea74b4742eb75bd8f055bb61e0b27e73ceae4aaf2a64dc52", "generation_seconds_batch": 17.26383608893957, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1556003093719482, "target_z": -1.1953124680013305, "ce": 6.23706579208374, "genuineness": 4.20609712600708, "identity": 0.3308030068874359, "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/casual_not_formal/89e87f3b3dfc9454da2c39be6785e17aac00fae3bc138c40f1ce927df2b333cc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "89f13c17628c2725a389d525084621dc207b0eef3d3dab5e39dba1cd8b4a6247", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "163fdcbefb3cc2b71e5a80aa862cb6ef3c5f226f1fee9078d9c1dad44b9d501e", "generation_seconds_batch": 24.04835745296441, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1057605743408203, "target_z": -1.3632812396848701, "ce": 6.2767791748046875, "genuineness": 4.457139015197754, "identity": 0.3033095896244049, "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/casual_not_formal/89f13c17628c2725a389d525084621dc207b0eef3d3dab5e39dba1cd8b4a6247.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "8a687b5ee2af484b779afb19a4ce249558a2a2fe402b280a36c4495afce14f99", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5c4d7b13d59a48bc37c59717ba4a710497b9394c46e756a46151f3360083657b", "generation_seconds_batch": 24.377253305050544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6420719623565674, "target_z": -0.3081054567997217, "ce": 6.324435234069824, "genuineness": 4.344170093536377, "identity": 0.36760517954826355, "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/casual_not_formal/8a687b5ee2af484b779afb19a4ce249558a2a2fe402b280a36c4495afce14f99.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "8aa74878cfa19810e815c8fb26fea907c58410e0790e3c997c18760eb4163220", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "61f58601459e79a9afc5406093266f29e850816ed83e7230af26dcbd6bbb975f", "generation_seconds_batch": 16.602567621972412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7867858409881592, "target_z": -1.7695313116666693, "ce": 6.395919322967529, "genuineness": 4.287685871124268, "identity": 0.5053185820579529, "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/casual_not_formal/8aa74878cfa19810e815c8fb26fea907c58410e0790e3c997c18760eb4163220.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "8abaa8088f8f9ef2a4bd294630092b8dbafd04a984ae08a3a6a7505376c86b43", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a7905c04146349a86beff02562b50791a54b73779c4e7da9f3bd4ac985c7381", "generation_seconds_batch": 24.430546098970808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.40917963638176486, "ce": 6.197352409362793, "genuineness": 4.400654315948486, "identity": 0.4286227524280548, "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/casual_not_formal/8abaa8088f8f9ef2a4bd294630092b8dbafd04a984ae08a3a6a7505376c86b43.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "8b66656c89fbd7da3140a84e5835aca64579439a61a4b5f0363ade84214e23f1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b93f2adcd9a5740200c3a2563b107087c6cfd7f09ec6f2b6ed5549e61585a524", "generation_seconds_batch": 22.944294171989895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -0.9570313205075228, "ce": 6.292664527893066, "genuineness": 4.143336772918701, "identity": 0.4493161737918854, "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/casual_not_formal/8b66656c89fbd7da3140a84e5835aca64579439a61a4b5f0363ade84214e23f1.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": "8ba3a3985cc974e04d19f428b44b24a7bf34ea7d484ecf6c89f5977ac7723475", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "7e86be02495d465c443912e7f16b924b4851469e8fcf2d8362f4a96ff8e36b3e", "generation_seconds_batch": 18.145177373080514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9630048274993896, "target_z": -0.7871093609591637, "ce": 6.117925643920898, "genuineness": 4.011539936065674, "identity": 0.6811990737915039, "hypothesis": "Die Schlüsse liegen auf dem Küchentisch. Ich suchte in der Schublade, dann neben dem Fenster. Nimm sie mit, wenn du gehst.", "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/casual_not_formal/8ba3a3985cc974e04d19f428b44b24a7bf34ea7d484ecf6c89f5977ac7723475.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "8bf6c216e46c71c6fdf10426010de09771308f2ede96ca82e0e086b0467ec033", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d6c88d075e63a8c65034e8541241467e3523e8c3851b23fad22b31dd5f2395f", "generation_seconds_batch": 25.484788287081756, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.29871129989624, "target_z": 0.23217777446327928, "ce": 6.268836498260498, "genuineness": 4.2751336097717285, "identity": 0.44569823145866394, "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/casual_not_formal/8bf6c216e46c71c6fdf10426010de09771308f2ede96ca82e0e086b0467ec033.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "8cad1214b8f7d9e476eb22f9f7c20393ee571072d1aec92f3bf6c1f4a233dcaa", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "49f01a57429f14f501efa3e085108cdd205eacb3819a57d87e8151b9aff07ab8", "generation_seconds_batch": 24.543578927987255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4265148639678955, "target_z": -0.41796874799988326, "ce": 5.919358730316162, "genuineness": 4.582659721374512, "identity": 0.625243067741394, "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/casual_not_formal/8cad1214b8f7d9e476eb22f9f7c20393ee571072d1aec92f3bf6c1f4a233dcaa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "8d1dc97e14b0bede902ba250e4c9a3a45979d8bd844f6ce36e3a85e984ccedf9", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f88d206fef53841748b20c9cccf02048a63ce11bf279aee8fecb22852353e00", "generation_seconds_batch": 17.986038436065428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6340622901916504, "target_z": -0.8085936613625662, "ce": 6.141753673553467, "genuineness": 4.162164688110352, "identity": 0.3728724718093872, "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/casual_not_formal/8d1dc97e14b0bede902ba250e4c9a3a45979d8bd844f6ce36e3a85e984ccedf9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "8db303120d1e7cff168169ad95e66853c1665592a375ed4ccae087eb33242954", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3de9a71190f282ff448004cdd6ff6a4d32d0effc0657a0f34e9a6f7f6f67a028", "generation_seconds_batch": 21.910004258039407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.876497507095337, "target_z": -1.4921874488437483, "ce": 5.228346347808838, "genuineness": 4.074300289154053, "identity": 0.5388078093528748, "hypothesis": "La Poripazi Ninacido D'Anjou not asindilas and the Sidi Musle de Neshdates and the Tatilas.", "wer": 0.9333333333333333, "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/casual_not_formal/8db303120d1e7cff168169ad95e66853c1665592a375ed4ccae087eb33242954.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "8e59c48e97164826fb1d41057316ba182f580ed03fcddcac9001d2dd0b06126b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "cdfe11c2b631bfa8b953c7ea2007b71d10670129252af15d640467cbcd3a6d29", "generation_seconds_batch": 17.95938637095969, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.639580011367798, "target_z": -0.12841795536826492, "ce": 5.967014789581299, "genuineness": 4.507347106933594, "identity": 0.7140485048294067, "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/casual_not_formal/8e59c48e97164826fb1d41057316ba182f580ed03fcddcac9001d2dd0b06126b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "8e84c6075aad8b4ed702e35a467a5334b77bf234146252cc4bcbe07dd4ae5d92", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "be635296a9a9e809d5bf0c224a3e8ac71b90c6d6fa4aea03bd386f80de6fa3f4", "generation_seconds_batch": 17.63638815190643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -0.8593749800916717, "ce": 6.300607204437256, "genuineness": 4.507347106933594, "identity": 0.589604914188385, "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/casual_not_formal/8e84c6075aad8b4ed702e35a467a5334b77bf234146252cc4bcbe07dd4ae5d92.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "8e8d9ab11daa3a36c21f4431fa2d01fb5049262161444a8b723e070b45b81720", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64d727fe10ee2de14073684dd0aadb0598102bca7f70abe7769960c176135da9", "generation_seconds_batch": 16.485142056946643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0758566856384277, "target_z": -1.2421875014310548, "ce": 6.141753673553467, "genuineness": 4.105680465698242, "identity": 0.6347880959510803, "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/casual_not_formal/8e8d9ab11daa3a36c21f4431fa2d01fb5049262161444a8b723e070b45b81720.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": "8e9145b97e52d6186af95fcf4e2c1e3e37534ba4031e005e3017109bcd32ae70", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "302d3ff6fb9a5c4cb968312e4d48cfa26286b508ee28dd923fdc8e194cc3395b", "generation_seconds_batch": 24.236826284090057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.356739044189453, "target_z": -0.39062505285703275, "ce": 6.086154937744141, "genuineness": 4.193545341491699, "identity": 0.6226407289505005, "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/casual_not_formal/8e9145b97e52d6186af95fcf4e2c1e3e37534ba4031e005e3017109bcd32ae70.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "8eb475cb4eeb136785549365f2ea14cc9b97f93ff081126386f75186b4f4dba9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "10d7f76961ec5ded73dbd059104b1c8228c7fe620a715f31495ba745e39c72b7", "generation_seconds_batch": 23.068853755947202, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773613691329956, "target_z": -0.7187499532681766, "ce": 6.125868320465088, "genuineness": 4.356721878051758, "identity": 0.6290931105613708, "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/casual_not_formal/8eb475cb4eeb136785549365f2ea14cc9b97f93ff081126386f75186b4f4dba9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "8fea17811c28f374766ed0e6a7d8c9bdefb4b701948244ac7e7a974b7fe4a536", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58676810b2e14221787ddab208736230dae4fc3b1db6ca802d5a46c0f14292b9", "generation_seconds_batch": 26.815919438027777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9929087162017822, "target_z": -0.6328124700566734, "ce": 6.086154937744141, "genuineness": 4.369274139404297, "identity": 0.503454864025116, "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/casual_not_formal/8fea17811c28f374766ed0e6a7d8c9bdefb4b701948244ac7e7a974b7fe4a536.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": "8fee1f6c0bc54fc9a60e05a38c6ceb623842e2245118319fda26c55cd152de5b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "11ff7724f9d4fe50b3806686b180a72609a02ae67e120aaf2b88a1d08b60c802", "generation_seconds_batch": 16.116155511001125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.28696346282959, "target_z": -0.6542968232434738, "ce": 6.221180438995361, "genuineness": 4.21237325668335, "identity": 0.6791661381721497, "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/casual_not_formal/8fee1f6c0bc54fc9a60e05a38c6ceb623842e2245118319fda26c55cd152de5b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "9014b961fe7fe5a943a0dd446ac0a2b5db3c9384cc9d6524a19d991b4188f2c2", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "244d00b8236b703737925f4c1526d25b8a317947d0595f90859d1b485a529212", "generation_seconds_batch": 17.240946538979188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9761770963668823, "target_z": -1.609374951123193, "ce": 6.324435234069824, "genuineness": 4.319066047668457, "identity": 0.5547148585319519, "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/casual_not_formal/9014b961fe7fe5a943a0dd446ac0a2b5db3c9384cc9d6524a19d991b4188f2c2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "9067b50b0d51a7cdaf8b158af019c5f0cb5da9702c0fa4e068121fe4693d0746", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b34f7fd1c13688f5a359faf394525cd44b83fe59f6d7036d768c66112999979e", "generation_seconds_batch": 25.347691197064705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9124526977539062, "target_z": 0.048095661470402944, "ce": 6.117925643920898, "genuineness": 4.507347106933594, "identity": 0.2731955945491791, "hypothesis": "Ik heb een kleine doosje gevonden. 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/casual_not_formal/9067b50b0d51a7cdaf8b158af019c5f0cb5da9702c0fa4e068121fe4693d0746.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "90e669fca4b2a4bbcc9cd3d976350e60ed78a8e888f8573ce26c200a27208d68", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "272679115984c811b60f81a627713831562d4d7cc734d647aeb9517ae0c5b992", "generation_seconds_batch": 16.250904027954675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.92313289642334, "target_z": -0.7109375239901994, "ce": 5.974957466125488, "genuineness": 4.432034492492676, "identity": 0.6971265077590942, "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/casual_not_formal/90e669fca4b2a4bbcc9cd3d976350e60ed78a8e888f8573ce26c200a27208d68.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": "9103a8ef7d88d593a92a28c7b94b5474ceaca20b22a4f615ceb63bf7f7ed2e0c", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "8723bd362df234dfe532f27218f8b626066f99621a86fd0b8625cd2aa9a82c0e", "generation_seconds_batch": 26.774112915969454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.539900302886963, "target_z": -0.45751959670960946, "ce": 6.324435234069824, "genuineness": 4.281409740447998, "identity": 0.772305965423584, "hypothesis": "Die Schlüsse liegen auf dem Küchentisch. Ich suchte in der Schublade, daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.14285714285714285, "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/casual_not_formal/9103a8ef7d88d593a92a28c7b94b5474ceaca20b22a4f615ceb63bf7f7ed2e0c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "91ac8d4e4980dde85da65fdc6c6fea7a9bf642087472de1b1093723e0ddebea9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "984487be8d06a37a23206af17846c8fec692659faae0e236af5642996c50ba84", "generation_seconds_batch": 21.911721925018355, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9784905910491943, "target_z": -0.07080074713916196, "ce": 6.268836498260498, "genuineness": 4.570107460021973, "identity": 0.5801808834075928, "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/casual_not_formal/91ac8d4e4980dde85da65fdc6c6fea7a9bf642087472de1b1093723e0ddebea9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "91be0683123c92cae6c1ccf5cd7cf67d46a795fd73a421ceab663f9b0a6b4ac4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "599897e558435af398aa6f4ceea5af4f884d5eb55e99f7c31446a40e20e96ada", "generation_seconds_batch": 18.194291541003622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.3203125920325918, "ce": 6.340320587158203, "genuineness": 4.369274139404297, "identity": 0.4936988055706024, "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/casual_not_formal/91be0683123c92cae6c1ccf5cd7cf67d46a795fd73a421ceab663f9b0a6b4ac4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "91cb5fb118559d39c99a9494129d63aa2f666f52964c19203113b535839371a7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "616dff6ecd6ea0758676ba1cbc433a73f2f8e498a0ccad78d15a932efa901e2c", "generation_seconds_batch": 26.823047554004006, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0758566856384277, "target_z": -1.507812490347574, "ce": 6.38797664642334, "genuineness": 4.381826400756836, "identity": 0.5088910460472107, "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/casual_not_formal/91cb5fb118559d39c99a9494129d63aa2f666f52964c19203113b535839371a7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "920b5153e50bf5ea3f276c234806309e9a19f2fc18820f0a5241a0d07cdf2352", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e7d559228045668c107f7efe8fccad331755d2221144d590e5b027d27aa57acd", "generation_seconds_batch": 18.756172602064908, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.032780647277832, "target_z": -0.46191401091995316, "ce": 6.09409761428833, "genuineness": 3.9173994064331055, "identity": 0.44707930088043213, "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/casual_not_formal/920b5153e50bf5ea3f276c234806309e9a19f2fc18820f0a5241a0d07cdf2352.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "9225f648177c0f249c4c5697731909a134c2322a44262bb31844b02b98d64a2d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adb7853c197f8a46fc2b22337b002ed5f45872d0f520ad6ecd7ce2d66ec827e3", "generation_seconds_batch": 17.884497439023107, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.619110345840454, "target_z": -1.1074217737631904, "ce": 6.006728172302246, "genuineness": 4.469690799713135, "identity": 0.679574191570282, "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/casual_not_formal/9225f648177c0f249c4c5697731909a134c2322a44262bb31844b02b98d64a2d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "923311d95cd9aafa0ab4273088cbabda14761005295ff1591be3a8f1408111ec", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d07748fad13ec1248a22c4a6cf07ce5037b83a9de56bc52a3364c8367ef533eb", "generation_seconds_batch": 25.70433899492491, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.888245105743408, "target_z": -0.6269531500226717, "ce": 6.300607204437256, "genuineness": 4.262581825256348, "identity": 0.2905834913253784, "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/casual_not_formal/923311d95cd9aafa0ab4273088cbabda14761005295ff1591be3a8f1408111ec.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "925c61034481845f52c7ec91d119e1e3c60167d2dc565227b7c3c4acf519b0a1", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b2ee7f4b686cdc4a27d63f5777c05dcd73c25e1d01d986aef2ff588b2e08539", "generation_seconds_batch": 17.241436044918373, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6290783882141113, "target_z": -1.072265630941922, "ce": 6.165581703186035, "genuineness": 4.331617832183838, "identity": 0.5577002763748169, "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/casual_not_formal/925c61034481845f52c7ec91d119e1e3c60167d2dc565227b7c3c4acf519b0a1.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": "928782ee8b1a415300c7c5fa19ab66a7ec62b1d23411bb33646d2986e7549bda", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 127, "seconds_generated": 10.16, "mp3_sha256": "09104065d4cfd602b2702d3574fc24d76aa209745d83b928e0f4bc0949fe0dd2", "generation_seconds_batch": 15.541306035942398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.22395133972168, "target_z": 0.2963866509923363, "ce": 5.927301406860352, "genuineness": 4.494794845581055, "identity": 0.4845530390739441, "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": 0.03999999999999915, "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/casual_not_formal/928782ee8b1a415300c7c5fa19ab66a7ec62b1d23411bb33646d2986e7549bda.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": "92a8f1de724983fde260892acb1117caf57355fab9472b3e003d449772853226", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "efe26866c8f424238b8537b48ee5ff726771d6a4426fa8b3216f29885f7d56e2", "generation_seconds_batch": 18.675916413078085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3667070865631104, "target_z": -0.5117187992009554, "ce": 6.308549880981445, "genuineness": 4.306513786315918, "identity": 0.4061884582042694, "hypothesis": "Die Schlüssel liegen auf dem Küchentisch. Ich suchte in der Schublade daneben dem Fenster. Nimm sie mit, wenn du gehst.", "wer": 0.09523809523809523, "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/casual_not_formal/92a8f1de724983fde260892acb1117caf57355fab9472b3e003d449772853226.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "92b2f7e3f265085d8e550591fc0edadf497dc7aa47d98f860c7b771f2eb71b0c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7e499c0835a8e15e92f34efbc89f941fecd9e6bbfbe38c656045c885a5c5ad0d", "generation_seconds_batch": 24.368041634908877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8266576528549194, "target_z": -1.449218769090352, "ce": 5.855817794799805, "genuineness": 3.8797430992126465, "identity": 0.44159573316574097, "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/casual_not_formal/92b2f7e3f265085d8e550591fc0edadf497dc7aa47d98f860c7b771f2eb71b0c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "9362fb7ddf238c5a5ada23944fec041df89befce6e708677f6479a0a04c9be2c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "911ab0ed346a0edec5a3d340194e0e4fb42020c5464279dcd6364806c7a63b85", "generation_seconds_batch": 16.641121591092087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.3359375223590515, "ce": 6.078212261199951, "genuineness": 4.055472373962402, "identity": 0.5700780153274536, "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/casual_not_formal/9362fb7ddf238c5a5ada23944fec041df89befce6e708677f6479a0a04c9be2c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "94419ba61a71eceae71efa4bb3b9527496f1befc6032b883a2ffc669bf851ba8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fa1ab55c9963f1f79e92cdaf831c54850562aac12fb12c2da4cfd7a1c3e01eef", "generation_seconds_batch": 19.731660658027977, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.3281249715933754, "ce": 5.792276382446289, "genuineness": 4.193545341491699, "identity": 0.5458816289901733, "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/casual_not_formal/94419ba61a71eceae71efa4bb3b9527496f1befc6032b883a2ffc669bf851ba8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "94717cecfcfcc244997ea2c3de1d1664f6430758bda3358637f00b61fb18f92d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5edbc700e21e25213320488331b7bcdf9893e72520db3fde1d0f6e3f7296aa6", "generation_seconds_batch": 17.589124253951013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4347031116485596, "target_z": -1.1367187306935493, "ce": 6.300607204437256, "genuineness": 4.444586753845215, "identity": 0.6572883129119873, "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/casual_not_formal/94717cecfcfcc244997ea2c3de1d1664f6430758bda3358637f00b61fb18f92d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "9481972aeffbbceba89cc62271b77b7c494211ff4222e927f52859df32764b33", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fea5118832e6f9eea27841e1262e8cc365d3005f79faf92c09b4c10a35e02ef", "generation_seconds_batch": 23.193395684007555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5436384677886963, "target_z": -0.39355466733235334, "ce": 6.205295085906982, "genuineness": 4.469690799713135, "identity": 0.7585093975067139, "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/casual_not_formal/9481972aeffbbceba89cc62271b77b7c494211ff4222e927f52859df32764b33.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "950245eee602742569849bd0bc72cfc2f94b7c8c3e72a2a915f2f68687a46229", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "242140028c72f81a07b9f1a1836d695350a465908d0328af1b9fda1438200883", "generation_seconds_batch": 26.26931932102889, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6451869010925293, "target_z": -0.29907224847914066, "ce": 6.086154937744141, "genuineness": 4.369274139404297, "identity": 0.6495987176895142, "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/casual_not_formal/950245eee602742569849bd0bc72cfc2f94b7c8c3e72a2a915f2f68687a46229.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "9559e3197d6ea2765704fa9b70d0f09d8515a0305961c2fcd2853020fd18fca6", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1e3f15bfa3d358052ffdbb487482f1d15be87ca53b0ee1e2cd698c8086b30c09", "generation_seconds_batch": 16.402832354069687, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5973947048187256, "target_z": -1.6679687508051684, "ce": 6.181467056274414, "genuineness": 4.2751336097717285, "identity": 0.3302483558654785, "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/casual_not_formal/9559e3197d6ea2765704fa9b70d0f09d8515a0305961c2fcd2853020fd18fca6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "9566afe82aae6b931703dca41f844b906af653b079abb97af18d437e59bef104", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da584a859c2c91b09424ff19ba19f213be931b50e7e7c16ca6be2be322e68b47", "generation_seconds_batch": 23.395032897009514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.452680826187134, "target_z": -0.28662107497019, "ce": 5.998785495758057, "genuineness": 4.168440818786621, "identity": 0.735512912273407, "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/casual_not_formal/9566afe82aae6b931703dca41f844b906af653b079abb97af18d437e59bef104.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": "95bb80c0d38a7fe3e90f21c6450876c32c89083a5fa7b6d51f22dd7fd415c5d5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0796e85c95017150399db02a3e3ac9f7df6e4e88b83018fa7089423caf57a931", "generation_seconds_batch": 26.774112915969454, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.977956771850586, "target_z": -0.865234394340066, "ce": 6.41180419921875, "genuineness": 4.406930446624756, "identity": 0.5936872363090515, "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/casual_not_formal/95bb80c0d38a7fe3e90f21c6450876c32c89083a5fa7b6d51f22dd7fd415c5d5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "95e3d9b4f6b1ecb635db874c3b9bc20cd1214ad37ad5a6e11e31c68050a87b44", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06ea425755ff863c2c67101bb201abd6a22329d69e181010d123322b32615ceb", "generation_seconds_batch": 23.290918337064795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.089384078979492, "target_z": 0.1303711275252836, "ce": 6.229123115539551, "genuineness": 4.1998209953308105, "identity": 0.33132636547088623, "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/casual_not_formal/95e3d9b4f6b1ecb635db874c3b9bc20cd1214ad37ad5a6e11e31c68050a87b44.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "975491cd75ed27dbf58d111bbd6caf9c284f7b1a1c8d8ca9769f6e7be7d8804d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "41c79690a2ce515f6db62daac76fbdfcc58bdfaec143ce2f4324198dd4921f03", "generation_seconds_batch": 20.984149588039145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.612168073654175, "target_z": -0.3178710967785886, "ce": 6.062326908111572, "genuineness": 4.344170093536377, "identity": 0.5778167247772217, "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/casual_not_formal/975491cd75ed27dbf58d111bbd6caf9c284f7b1a1c8d8ca9769f6e7be7d8804d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "975ae7f4d8ae44dc6c027869710dc294c59d63e27bc59a1bf81a4bd0a1abec5e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b990f314ae2638e2a3f78a2aa1a8f7ea6e18fa70bfcc648ed3085329ca493aa7", "generation_seconds_batch": 21.248646283056587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.3632812677409345, "ce": 5.982900142669678, "genuineness": 4.118232727050781, "identity": 0.7178045511245728, "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/casual_not_formal/975ae7f4d8ae44dc6c027869710dc294c59d63e27bc59a1bf81a4bd0a1abec5e.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": "97969840df2b80152f0532348ab96a4da697e99204ea9fdfdd6fe736f916a68b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ae237e4c9669c666855069a4e87a425a43a95acc7a414efa41c22cf8ca502013", "generation_seconds_batch": 26.220426419051364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.741128444671631, "target_z": -0.3024902293258868, "ce": 6.062326908111572, "genuineness": 4.632867813110352, "identity": 0.8098049759864807, "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/casual_not_formal/97969840df2b80152f0532348ab96a4da697e99204ea9fdfdd6fe736f916a68b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "97a39aab0cfb13c4e06ccc451014edcb90317973d3facd7420114cee88321898", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b72be8c4781dce9db917a4c77cbb9676e3b18807c6eb39aebfbea39c0bf1542e", "generation_seconds_batch": 18.830436743912287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5941903591156006, "target_z": -0.9414062654326756, "ce": 5.919358730316162, "genuineness": 4.2374773025512695, "identity": 0.655630350112915, "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/casual_not_formal/97a39aab0cfb13c4e06ccc451014edcb90317973d3facd7420114cee88321898.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "97b934149449901a79353dde253a63a98085758b75043e966a3832700de0247f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0710c738714c8acfbf724f321623760a2fc79c100047391617eef4b77d3e48ba", "generation_seconds_batch": 22.549734112923034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2072198390960693, "target_z": -0.32714841478598694, "ce": 6.046441555023193, "genuineness": 4.193545341491699, "identity": 0.4467731714248657, "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/casual_not_formal/97b934149449901a79353dde253a63a98085758b75043e966a3832700de0247f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "97bfce16a8fa84e9960990ec5e5a657de0b4db0a2c3089b8846f88cb623a63c6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52d23ec94ff6d32adc5821701a2ef1e88794312cbd0bcb2aa3d7f7908ec3595b", "generation_seconds_batch": 21.95025450293906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7985334396362305, "target_z": -0.7597656436981732, "ce": 6.3641486167907715, "genuineness": 4.595211505889893, "identity": 0.35690730810165405, "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/casual_not_formal/97bfce16a8fa84e9960990ec5e5a657de0b4db0a2c3089b8846f88cb623a63c6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "9843f0dbcb75822ed662fad8f110a5c563803e4a8fe9c8bb3b24b86b49e4cf3c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88b6e22f9d14e57818bb30872a72fe166e682c763a47e753db2d50534c79a5a7", "generation_seconds_batch": 19.194249292020686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6871063709259033, "target_z": -1.8085937700603945, "ce": 6.332377910614014, "genuineness": 4.042920112609863, "identity": 0.6220777034759521, "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/casual_not_formal/9843f0dbcb75822ed662fad8f110a5c563803e4a8fe9c8bb3b24b86b49e4cf3c.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": "985cf1aaca8279e0d00eb8dfe86a921e2652eae9f6c06e25d9ab1d4810e05fc4", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ac2f9a04573ccb4788b3472f10ebc3b74aa758590dc40eb5aa1318ff64f7f010", "generation_seconds_batch": 23.568580292048864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8982131481170654, "target_z": -1.076171900061088, "ce": 6.078212261199951, "genuineness": 4.337893962860107, "identity": 0.7230904698371887, "hypothesis": "Ik schreef het nummer op de envelop, lees het goed en controleer de laatste twee cijfers. Bel me wanneer je even tijd hebt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/985cf1aaca8279e0d00eb8dfe86a921e2652eae9f6c06e25d9ab1d4810e05fc4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "98682f46257d44896154b06ffccb3bbcce054c175875d3daeb712c575e6705c1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6affe42e524dfcf6d463e32367b40764dbe6b4a669b51275cc303e626fa4a603", "generation_seconds_batch": 17.71769471198786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.933100938796997, "target_z": -0.6523437500144071, "ce": 6.197352409362793, "genuineness": 4.13078498840332, "identity": 0.13166549801826477, "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/casual_not_formal/98682f46257d44896154b06ffccb3bbcce054c175875d3daeb712c575e6705c1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "9897f82d58bfc2208ee065351d61e99013dd90c26520b3aa320c01ee957fa60f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86b06403147a735ba10b3e2f473a9e17c75f11badd61cfadc73396520e13a9dc", "generation_seconds_batch": 23.511907604988664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8526451587677, "target_z": 0.13574220810704812, "ce": 6.1020402908325195, "genuineness": 4.507347106933594, "identity": 0.46721503138542175, "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/casual_not_formal/9897f82d58bfc2208ee065351d61e99013dd90c26520b3aa320c01ee957fa60f.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": "98b9524d15e13fba7573f42abb1aedb44dbbcb06102ec0d9f488fba2b9a09998", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bb529043f99f8c6cea36d14420869db1679bf29369546a1274f14a91ceb692a4", "generation_seconds_batch": 25.396506458055228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1057605743408203, "target_z": -1.5507812249068955, "ce": 6.372091293334961, "genuineness": 4.268857479095459, "identity": 0.7392871379852295, "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/casual_not_formal/98b9524d15e13fba7573f42abb1aedb44dbbcb06102ec0d9f488fba2b9a09998.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "98e323870bfd3f13f75c846b3d7f647c3448711c3d7c08208096f0e8283ac53f", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d944676e61f87fc1d2cab4dae5e4a49a809bb6b46b432d9300bf2a88347cd826", "generation_seconds_batch": 18.648007082985714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.806721806526184, "target_z": -1.5507812938056809, "ce": 6.086154937744141, "genuineness": 4.325342178344727, "identity": 0.5880565643310547, "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/casual_not_formal/98e323870bfd3f13f75c846b3d7f647c3448711c3d7c08208096f0e8283ac53f.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "6b2fb5e1c3985de0cb07b22b0d5a7a3b042b368f0cc041277dad3b845f70e0a8", "generation_seconds_batch": 17.828491030959412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.396610975265503, "target_z": -0.47265629761462796, "ce": 6.372091293334961, "genuineness": 4.419482707977295, "identity": 0.41007697582244873, "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/casual_not_formal/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "9abacb6f4a2e4a81f942e80bee75f3a418091ac5ab1e54a61f0c0fc21378a9a9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "750105670fc96762e0e64cbb772f98345bea8fc662e97aba3b96b0ff2952f2f3", "generation_seconds_batch": 24.688993592979386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9436025619506836, "target_z": 0.041015602168932, "ce": 6.213237762451172, "genuineness": 4.545003414154053, "identity": 0.7918453216552734, "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/casual_not_formal/9abacb6f4a2e4a81f942e80bee75f3a418091ac5ab1e54a61f0c0fc21378a9a9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "9adaae0c25b7092a3d4e21f7654a6705bd57e7fc096353821f726806933ab1cb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "399b98d1be111d8ff7a4d0386d46d7b77e70db6a7ce6e1ec1358ecca12e48648", "generation_seconds_batch": 22.7830135790864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7803773880004883, "target_z": -0.18749998548294483, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.7832396030426025, "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/casual_not_formal/9adaae0c25b7092a3d4e21f7654a6705bd57e7fc096353821f726806933ab1cb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "9b286cc39a093b5737e8fa7c8cc9eb5367795d74f5b6ad5c12d7438fdfb4c096", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "755b2b15f6a9431bdb90f619ba8ebcc8735d18ca09e32b5cebe3fa0a8b829a6d", "generation_seconds_batch": 26.499635227024555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446450710296631, "target_z": -0.4316406717126152, "ce": 5.990842819213867, "genuineness": 4.494794845581055, "identity": 0.6954542398452759, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce que 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/casual_not_formal/9b286cc39a093b5737e8fa7c8cc9eb5367795d74f5b6ad5c12d7438fdfb4c096.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "9b2d95f3e7b4b85207044c656e255cc55ee06a8f00f5cd7b20f6d8d7940cccae", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c738087914c648707c26f182f14ab7de4fb31acfc442547a4a15bcb3e801ce42", "generation_seconds_batch": 24.420667428988963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7387256622314453, "target_z": -0.8574219273504007, "ce": 6.221180438995361, "genuineness": 4.469690799713135, "identity": 0.5251806378364563, "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/casual_not_formal/9b2d95f3e7b4b85207044c656e255cc55ee06a8f00f5cd7b20f6d8d7940cccae.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "9b6f1e896cff683e995672d39a8c2547d82638c95564085c95b35b5cae7f0659", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6ae9b6b00b9c878a17006595fe67a74612748e9b07df987c931806d717f15031", "generation_seconds_batch": 17.148230073973536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.121779918670654, "target_z": 0.2548827932063553, "ce": 6.006728172302246, "genuineness": 4.796044826507568, "identity": 0.7845262885093689, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/9b6f1e896cff683e995672d39a8c2547d82638c95564085c95b35b5cae7f0659.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": "9b726c0484bc02191f3ccc533665c10bce171a148b34e8e86b307638be866f72", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4166a0a947a77f89c98b2231affa99231054213622d621a228a3b37bb6bc3157", "generation_seconds_batch": 25.396506458055228, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.729135751724243, "target_z": -0.2798461885972636, "ce": 6.372091293334961, "genuineness": 4.607763767242432, "identity": 0.4806089699268341, "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/casual_not_formal/9b726c0484bc02191f3ccc533665c10bce171a148b34e8e86b307638be866f72.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "9b7bc20311bd901f165301f0cd7266ec704a3d53c87f3288fb5a7f42ced9230b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2bf00d88c98c6182923bad4e14330497c5bb166d61d6cc4f774e090a477b37a9", "generation_seconds_batch": 24.447202009963803, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6215131282806396, "target_z": -0.23999017918449833, "ce": 6.078212261199951, "genuineness": 4.570107460021973, "identity": 0.7893752455711365, "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/casual_not_formal/9b7bc20311bd901f165301f0cd7266ec704a3d53c87f3288fb5a7f42ced9230b.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": "9bcc967319e5da4259ef8aeff34eb96f695d6458f49d341428a676cafb07132a", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "60e25dd89ab78cfec2a3c93aeaf585a0e548ad53248f31672930202290b84d14", "generation_seconds_batch": 22.76763163297437, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2652478218078613, "target_z": -1.484375021653126, "ce": 6.260893821716309, "genuineness": 4.030367851257324, "identity": 0.8377460241317749, "hypothesis": "Las llaves están en la mesa de la cocina, mira 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/casual_not_formal/9bcc967319e5da4259ef8aeff34eb96f695d6458f49d341428a676cafb07132a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "9bef0a2676b35ae53ba4f5474403696595e0e11f9a053cd53fb6d29d1c1d2c79", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d0e2b5f13b28ca61bde432122d6b745e8a33d479d7a72e0351eaa37b96a39c7", "generation_seconds_batch": 24.937223019078374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.473150253295898, "target_z": 0.36328117685366557, "ce": 6.0305562019348145, "genuineness": 4.6705241203308105, "identity": 0.7537895441055298, "hypothesis": "J'ai trouvé la petite ditée 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/manual-comparison-v1/audio/casual_not_formal/9bef0a2676b35ae53ba4f5474403696595e0e11f9a053cd53fb6d29d1c1d2c79.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "4c1be37a589a2eafba169284f4dda69b6a6efafd6f2869823d43cb0c4caffd4c", "generation_seconds_batch": 16.119336390984245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.483830690383911, "target_z": -0.41699217974124236, "ce": 6.300607204437256, "genuineness": 4.413206577301025, "identity": 0.6782809495925903, "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/casual_not_formal/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "f7054c2fcb01c3a3091d00d0bddc6da61fb2a0e9f3cff68fe1d88c49b967c7bd", "generation_seconds_batch": 15.968530033016577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2620434761047363, "target_z": -0.44531252150265876, "ce": 5.784333229064941, "genuineness": 4.570107460021973, "identity": 0.5808717012405396, "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/casual_not_formal/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "9d3d7c01c68419fd82c300d293b59da5d87585fc1f3076bdcc1b05614e8e14e8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ed911a5d5f6a854c29fd383ae6be784c12f67fb8597af450496954557ad06e71", "generation_seconds_batch": 24.102148448000662, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -1.1015624610365382, "ce": 5.677107334136963, "genuineness": 4.281409740447998, "identity": 0.5774948596954346, "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/casual_not_formal/9d3d7c01c68419fd82c300d293b59da5d87585fc1f3076bdcc1b05614e8e14e8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "9def3c9eb340601eb92b1481a20f5014e64dfe0827895c9af09444fd7b8f97d6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "88480faf92fbbda075918eae2f32b978680a352290bea682aaab1086723a6cef", "generation_seconds_batch": 26.061494419001974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.686304807662964, "target_z": 0.17211915502400132, "ce": 5.919358730316162, "genuineness": 4.231201171875, "identity": 0.7369040250778198, "hypothesis": "Ik heb het kleine cadeje gevonden. Het stond naast het oude raam. We kunnen er samen in.", "wer": 0.1111111111111111, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/9def3c9eb340601eb92b1481a20f5014e64dfe0827895c9af09444fd7b8f97d6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "9df32c5707f567f6e12ffb29cd155f6af6b94f050e8cc6de416a6de0c802711f", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0fbf8f45f2a4c6ceeda16ded7538e83e1d1064ab2e8cfd9be827fb007150f3cf", "generation_seconds_batch": 16.821575233945623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.077636241912842, "target_z": -0.7792969100848854, "ce": 6.165581703186035, "genuineness": 4.381826400756836, "identity": 0.72670578956604, "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/casual_not_formal/9df32c5707f567f6e12ffb29cd155f6af6b94f050e8cc6de416a6de0c802711f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "9f183f65bbd54a6f3459158ad76167ed71e747a3333bcd28c7795d87a6228701", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1529ae2c78aae6d79a67c707ed9b8aded08c5a754cbe897f1d96b6791c4c3755", "generation_seconds_batch": 16.099584136041813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.98614501953125, "target_z": -1.4804688032948312, "ce": 5.804190158843994, "genuineness": 4.105680465698242, "identity": 0.6831617951393127, "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/casual_not_formal/9f183f65bbd54a6f3459158ad76167ed71e747a3333bcd28c7795d87a6228701.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "9f591046c02be88a93b4737c9034212355da8e2d4b18917a7bf69c55ea093b04", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "93ea08b292d467867e82b56c67d3656920d10067483930e37258f74e37473d35", "generation_seconds_batch": 25.618976086960174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.42331075668335, "target_z": 0.42382816100850884, "ce": 6.141753673553467, "genuineness": 4.400654315948486, "identity": 0.5609644651412964, "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/casual_not_formal/9f591046c02be88a93b4737c9034212355da8e2d4b18917a7bf69c55ea093b04.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": "9fb3c431e6af5bd431ed1b9236873b04c2f28a58b88384921f635b5316981b94", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ef2bf010c1e73e0fe6ce780cd6bda4417e1244c2501d06ee75597e43a8574172", "generation_seconds_batch": 18.653824200038798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.987924814224243, "target_z": -0.6660156610037038, "ce": 6.332377910614014, "genuineness": 4.20609712600708, "identity": 0.37982529401779175, "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": 0.03999999999999915, "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/casual_not_formal/9fb3c431e6af5bd431ed1b9236873b04c2f28a58b88384921f635b5316981b94.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": "9fc94bd50c9082d9cc12308b4db74679ab2c651a4bdf0845439a52128ef9a416", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ee556fe58df58fb7f52658d4a50f5c08bea1e6ccddb2964550da10e2296414d8", "generation_seconds_batch": 23.568580292048864, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.047732353210449, "target_z": -0.8457031570530957, "ce": 6.038498878479004, "genuineness": 4.243753433227539, "identity": 0.7090398073196411, "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/casual_not_formal/9fc94bd50c9082d9cc12308b4db74679ab2c651a4bdf0845439a52128ef9a416.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "a00fce694760913c480a3241dbbb92fc2273732a4bd2549ff398a4253fb6047b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca30c749aa2fd0ba932981391c25d762066d96b57ab184fdfddde22dffd1d9f4", "generation_seconds_batch": 23.887470773071982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3848633766174316, "target_z": -1.351562528238544, "ce": 6.324435234069824, "genuineness": 4.15588903427124, "identity": 0.5196438431739807, "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/casual_not_formal/a00fce694760913c480a3241dbbb92fc2273732a4bd2549ff398a4253fb6047b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a01b484f2c03ffef3f35ad538d817d82b8cb12c0033dc6b94e15e1830a8fb016", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0815adc9141ab4f5c38c0deb8a8476f1c2a25cceb4ee2bb16b4ee09381a8b98d", "generation_seconds_batch": 16.854664427926764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866429328918457, "target_z": -0.42968755496519895, "ce": 6.252951145172119, "genuineness": 4.406930446624756, "identity": 0.5022404193878174, "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/casual_not_formal/a01b484f2c03ffef3f35ad538d817d82b8cb12c0033dc6b94e15e1830a8fb016.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "a01e06f2ef20f7ee9bd668b0f87eab8caa6e11559f14a324df47e0de984e097a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1202b5de343ff4fabd54b1aab76f2ecf14b97397f11447f9cb85765dcbc6e8fa", "generation_seconds_batch": 17.373828989919275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.259551525115967, "target_z": -0.4580078254131206, "ce": 6.117925643920898, "genuineness": 4.444586753845215, "identity": 0.47267991304397583, "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/casual_not_formal/a01e06f2ef20f7ee9bd668b0f87eab8caa6e11559f14a324df47e0de984e097a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "a0257f69397fc9751793ce6713d2119f2eff6a0790e030ef8b5466aab7da21ea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "75de3cfa23c419640208db6195ae1c345d26ee89b99f80588fb7a277a4fb9d80", "generation_seconds_batch": 22.530166022013873, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.87827730178833, "target_z": -0.8320312568279287, "ce": 6.078212261199951, "genuineness": 4.331617832183838, "identity": 0.5558862090110779, "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/casual_not_formal/a0257f69397fc9751793ce6713d2119f2eff6a0790e030ef8b5466aab7da21ea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "a04eb22b4b8517373ef516ffe364cb2c863df3393b1f4950b5095371dc8b7a2d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff8c2eabbf32de2b0b2a695e639b54d39e308bc77455c0208fbc2b16d45f1069", "generation_seconds_batch": 17.61271170398686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -1.3007812618358257, "ce": 6.372091293334961, "genuineness": 4.168440818786621, "identity": 0.553987979888916, "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/casual_not_formal/a04eb22b4b8517373ef516ffe364cb2c863df3393b1f4950b5095371dc8b7a2d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a0bbde734ce95031d3a60f11015edae16b9d1b6043df1e6d97ba6a0af14fa4c4", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f12b41afe9f0ccc9b01fa98bcc6ff17e17f49ed85ef82ed07d5b727a9e68c6c", "generation_seconds_batch": 17.82794892298989, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3218512535095215, "target_z": -0.4902343725484772, "ce": 6.356205940246582, "genuineness": 4.381826400756836, "identity": 0.45901253819465637, "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/casual_not_formal/a0bbde734ce95031d3a60f11015edae16b9d1b6043df1e6d97ba6a0af14fa4c4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "a0c6ad3c71020051696f9c07dc7a227a5fd04d4a540e2fc827f6ad8c2783489c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b81ef912cba6e208fa2a8591c63ef5e1d31517e8f4b48dc7f2e5a0efaf15282a", "generation_seconds_batch": 26.539110217010602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -0.863281210455472, "ce": 6.197352409362793, "genuineness": 4.369274139404297, "identity": 0.6519124507904053, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd, zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/a0c6ad3c71020051696f9c07dc7a227a5fd04d4a540e2fc827f6ad8c2783489c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "a0eea57bbf1926c72ce1f6d8f86552702f9f11a7897a9764b16c70fc80a5256e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01216089ded5562f66cd397ce2a706d29327e6b4d33d3f71ca0b12300af3719a", "generation_seconds_batch": 23.709864328964613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9822285175323486, "target_z": 0.16259764549607023, "ce": 6.046441555023193, "genuineness": 4.174716949462891, "identity": 0.4785504639148712, "hypothesis": "Die Tür ist noch offen. Ich habe dein 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/casual_not_formal/a0eea57bbf1926c72ce1f6d8f86552702f9f11a7897a9764b16c70fc80a5256e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "a0f4d7b2ff36bcbd961c4e4c95eb65c44da2265d9c9f8bd5868a71ffd0672316", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b3797fdac4c84264428e4b87a171c4008ae5f81cfec6ffabc673478929eb6c4", "generation_seconds_batch": 24.41993298009038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.135664463043213, "target_z": -1.414062439146502, "ce": 5.982900142669678, "genuineness": 4.293961524963379, "identity": 0.7320353388786316, "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/casual_not_formal/a0f4d7b2ff36bcbd961c4e4c95eb65c44da2265d9c9f8bd5868a71ffd0672316.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "a14b56c8b747c93727e708d42745d0ed4423162d9096352911be8c2ab9727866", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c8a5af31d8b687e06d4871602d154550ab0dc4d6b5e17ba4fddf87b0f369930", "generation_seconds_batch": 21.84458421205636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.3007813120748581, "ce": 6.340320587158203, "genuineness": 4.068024158477783, "identity": 0.6390777826309204, "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/casual_not_formal/a14b56c8b747c93727e708d42745d0ed4423162d9096352911be8c2ab9727866.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "a1733a333150865869a25a70b0f26a6bb1a7fc4c01ac585ebb420cb7dc8a10f0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6724fb4a3bcd56c576c67221a6fe140e24b8cce62801152fd9567294d6b8de94", "generation_seconds_batch": 23.831461454974487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.285183906555176, "target_z": -1.4218748462415112, "ce": 5.974957466125488, "genuineness": 3.9173994064331055, "identity": 0.5251245498657227, "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/casual_not_formal/a1733a333150865869a25a70b0f26a6bb1a7fc4c01ac585ebb420cb7dc8a10f0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "a210a3dc645351212991076b5faa10dd0d597b5a4553ce0cbae564909722d9a6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "48d5830441fc9b185ee77d19d46187cdd48d83be1640a6b9b99376ecfebdef4a", "generation_seconds_batch": 22.495720722014084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6458098888397217, "target_z": -0.009277345492199503, "ce": 6.221180438995361, "genuineness": 4.256305694580078, "identity": 0.37556636333465576, "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/casual_not_formal/a210a3dc645351212991076b5faa10dd0d597b5a4553ce0cbae564909722d9a6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "a261fe1198c4bf8e7506a2a161d3ede4d1a04aae87c7fac1fcf79157c492bffb", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5dee9cfaecdc2339d3a4ff57d302a7fe74ee698dbeb23a5bc9cc77ce1c70a21", "generation_seconds_batch": 23.6127446340397, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -1.0468749653136973, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.4471808075904846, "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/casual_not_formal/a261fe1198c4bf8e7506a2a161d3ede4d1a04aae87c7fac1fcf79157c492bffb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a2c5d57091023e2c1a50c4181609a364434fa1813636c508e6aaf48a27727346", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "951045ee21c47982002e3357748bf59c97cafa5ab536fa190672040400fc5f3a", "generation_seconds_batch": 23.06136308494024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8594980239868164, "target_z": -0.21899415172088676, "ce": 6.403861999511719, "genuineness": 4.570107460021973, "identity": 0.4202573895454407, "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/casual_not_formal/a2c5d57091023e2c1a50c4181609a364434fa1813636c508e6aaf48a27727346.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "a30ba1bf645e14ea7dd36ba0f9e17112d25ea39f951e8cc3e2445e3d48886b60", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "43cdaae2c16cfbb11ed2b258885931924629d593fb7dfe53bc08c81ad8122c77", "generation_seconds_batch": 16.529483002028428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.1875000991398483, "ce": 6.046441555023193, "genuineness": 4.444586753845215, "identity": 0.7187227606773376, "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/casual_not_formal/a30ba1bf645e14ea7dd36ba0f9e17112d25ea39f951e8cc3e2445e3d48886b60.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "a3d1343ed1f662ef12e2d2598d00e7fc6ace7ec50ba79ccbb456faab002c6eb6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "524865bb9d5f75dc28d8d83860e90949c2413fafc47b1fbdad785af811478df3", "generation_seconds_batch": 24.005818294011988, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.04274845123291, "target_z": -0.5351562371653171, "ce": 5.903473377227783, "genuineness": 4.532451152801514, "identity": 0.7868440747261047, "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/casual_not_formal/a3d1343ed1f662ef12e2d2598d00e7fc6ace7ec50ba79ccbb456faab002c6eb6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "a40698c75a961630f14584cdb2a48d6a7a9935c4c8666cbb33530b5ec0522f14", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3da91060faeb05c6bbb45040f31a7c400e547f4f482556ade5dcb9eaa0b39147", "generation_seconds_batch": 21.45494276296813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.875072956085205, "target_z": 0.0834960657137474, "ce": 5.919358730316162, "genuineness": 4.005263805389404, "identity": 0.1651293784379959, "hypothesis": "Waarden de deur staat nog open? Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/a40698c75a961630f14584cdb2a48d6a7a9935c4c8666cbb33530b5ec0522f14.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": "a459358d705ec5e6e10f59b4a7a179b71d6c8ced886370b78a3460bc8138fe31", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4f3dcb97c79b76399b27495ad9b3a351eb0b20a576e3ec559b6ebdf3121a14a1", "generation_seconds_batch": 24.031974275945686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.6875000207150956, "ce": 6.1894097328186035, "genuineness": 4.400654315948486, "identity": 0.4815707802772522, "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/casual_not_formal/a459358d705ec5e6e10f59b4a7a179b71d6c8ced886370b78a3460bc8138fe31.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a47962b43b556e3726a01b85a518c5a4eb9ce9a94f0e9c1e592e825ed30767bf", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9a45f77effd72cc83f32abe7b0adac47f8a8e26506dcf0501c302cf4da87d4d", "generation_seconds_batch": 25.828512109001167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.007148265838623, "target_z": -0.07031258599742515, "ce": 6.348263263702393, "genuineness": 4.595211505889893, "identity": 0.27962011098861694, "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/casual_not_formal/a47962b43b556e3726a01b85a518c5a4eb9ce9a94f0e9c1e592e825ed30767bf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "a4c039a7b6192badffa0fe97e4e6f5ce6a016f37d156595af6b9720ff5821e7a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23206b8956a4c89b9c645fafea3061403c4532d213ec26df1ba517bc6236bdd0", "generation_seconds_batch": 22.042742638965137, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.627120018005371, "target_z": -0.05566404660931547, "ce": 6.252951145172119, "genuineness": 3.8797430992126465, "identity": 0.2611837387084961, "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/casual_not_formal/a4c039a7b6192badffa0fe97e4e6f5ce6a016f37d156595af6b9720ff5821e7a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "a513c9dd936272d6c679b22369992f20a73475bc508a7c8e34ac5027042c25fa", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9e56e667fdd9dd10436ef879d5c812220bc3698658d1aafdc7adcc6055225c57", "generation_seconds_batch": 17.61271170398686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9562411308288574, "target_z": -1.5156249542062072, "ce": 6.467403411865234, "genuineness": 4.362998008728027, "identity": 0.7056628465652466, "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/casual_not_formal/a513c9dd936272d6c679b22369992f20a73475bc508a7c8e34ac5027042c25fa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "a590480928cfb3d7cf55e23c9fc99e1cfb88f9cd98fbd3dd2eaf03b7069a1866", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f1b53dd8ecf5b9bd8f0939db813e01b7b46cc6b2f31c7422c0134e7ee47dc9de", "generation_seconds_batch": 21.491997733945027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.245311737060547, "target_z": -1.3671875585212299, "ce": 6.062326908111572, "genuineness": 4.20609712600708, "identity": 0.6971740126609802, "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/casual_not_formal/a590480928cfb3d7cf55e23c9fc99e1cfb88f9cd98fbd3dd2eaf03b7069a1866.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "a663176f89603e1d4a19e439d7bad8608b4f30f47bd1b1986447b849ee8ce8ec", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ecc7657b86202ff33ce8a1489820e8e701f452e8242697cfc4f61a32887fb73", "generation_seconds_batch": 23.91885087499395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5087504386901855, "target_z": -0.24316408045543247, "ce": 6.086154937744141, "genuineness": 4.507347106933594, "identity": 0.5380886793136597, "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/casual_not_formal/a663176f89603e1d4a19e439d7bad8608b4f30f47bd1b1986447b849ee8ce8ec.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "a694a9225d9b4af925804efb1d9933ed058316a29d0e484ee8fe5d0ae60250c4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "a29d8b1ebd8bac14aac60f27eff2ee271fba1e178ce289575b13848384a4afef", "generation_seconds_batch": 23.37159023899585, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1798079013824463, "target_z": -0.42675779603645314, "ce": 5.6254801750183105, "genuineness": 4.2249250411987305, "identity": 0.456535279750824, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás linda.", "wer": 0.06666666666666667, "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/casual_not_formal/a694a9225d9b4af925804efb1d9933ed058316a29d0e484ee8fe5d0ae60250c4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a7674a1ba4a4fce1dc8904755773d0e05764a6f838ea6f2490816bcd2b12938c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6f3ebe9d8bd04e46666b96a54130604785887565468fbac241f8ae91afd56956", "generation_seconds_batch": 17.213931355974637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.34356689453125, "target_z": 0.3671874530192618, "ce": 6.2767791748046875, "genuineness": 4.406930446624756, "identity": 0.4915146827697754, "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/casual_not_formal/a7674a1ba4a4fce1dc8904755773d0e05764a6f838ea6f2490816bcd2b12938c.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": "a807c2209fd5e8447727f33799eaf41fc0a8635f813f03f071f410d87dff3512", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a9c97776da09f71f91e15c6379bcf1de808bff39a59b2be65ab7a3892eab8bb", "generation_seconds_batch": 17.243588243960403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.919928789138794, "target_z": 0.04211427859667805, "ce": 6.1020402908325195, "genuineness": 4.457139015197754, "identity": 0.5114941596984863, "hypothesis": "De weg gaat linksaf naar de brug, loep verder tot aan het kleine houtenhek. Ik wacht aan de andere kant.", "wer": 0.19047619047619047, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/a807c2209fd5e8447727f33799eaf41fc0a8635f813f03f071f410d87dff3512.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "a83a0f9bf64d64233827f7db952d8ccac8ff7c29ff3f317f18ab4639879f1012", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "975768fb70adfcac5535d0b800d18d375e8b636a5fb73bb1c396cf95f642115a", "generation_seconds_batch": 26.607085130061023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.263823509216309, "target_z": 0.3320313285984341, "ce": 6.141753673553467, "genuineness": 4.68307638168335, "identity": 0.6028422117233276, "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/casual_not_formal/a83a0f9bf64d64233827f7db952d8ccac8ff7c29ff3f317f18ab4639879f1012.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a85137527e2f4cc4200e3d9485188178b3e03c63febe63a0480155c3781df656", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "98815d90508c920946a841a5c3680b8623ad1f45ee69a485a442f7d8baa80fab", "generation_seconds_batch": 17.213931355974637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.827413558959961, "target_z": -0.06408693407967803, "ce": 6.117925643920898, "genuineness": 4.469690799713135, "identity": 0.4553060531616211, "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/casual_not_formal/a85137527e2f4cc4200e3d9485188178b3e03c63febe63a0480155c3781df656.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "a87441ff8f7ee362db19a3c972fdbcfad2a7fba5eec1bd4447d11492f51e4b2e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1593eafe98e4593a4d5711bc18d2693e61ef71b801dfab75cd6da80e3847bf09", "generation_seconds_batch": 18.895433694007806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.149191856384277, "target_z": 0.17211914149389912, "ce": 6.308549880981445, "genuineness": 4.570107460021973, "identity": 0.35239532589912415, "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/casual_not_formal/a87441ff8f7ee362db19a3c972fdbcfad2a7fba5eec1bd4447d11492f51e4b2e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "a8a0fb45a55f4bb30ab1ad532a46c227ebb0213704f340f8a910dc571754fa00", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dc481f9556ecb21b6652e7241fc6679beeb43918f0e612fec0ab5fe8ecec50f9", "generation_seconds_batch": 17.931166169932112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2652478218078613, "target_z": -1.2148438002854518, "ce": 6.268836498260498, "genuineness": 4.1998209953308105, "identity": 0.8747777342796326, "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/casual_not_formal/a8a0fb45a55f4bb30ab1ad532a46c227ebb0213704f340f8a910dc571754fa00.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "22c2159fe1d30cdd1a1d90c1e8a6cd2e313cc993b013e9b85bee6f11d6c1c442", "generation_seconds_batch": 16.991411286988296, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.7558593334089319, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.7508885264396667, "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/casual_not_formal/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "a8c35657c025846df3b399e7b35e40e8f1f24055476e133b4452f7dbf9b1f12e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3f1522b0a8ddcae95d7f0243cccc7e4fe3ae90a110cd6827a31803ffc08c1a70", "generation_seconds_batch": 19.280610741931014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.4453125954901753, "ce": 6.0305562019348145, "genuineness": 4.287685871124268, "identity": 0.4370208978652954, "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/casual_not_formal/a8c35657c025846df3b399e7b35e40e8f1f24055476e133b4452f7dbf9b1f12e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "a8efbf50b855b1ff6f8ab1e3875b6127c587ffc4fbd84758006854ba30aa5f84", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7dd2ec4a3a9faeefc090306fd87dc92879dfe2ec17b98876ab49bfc3752aa4ad", "generation_seconds_batch": 24.573775270953774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.159872055053711, "target_z": -0.6103515568974309, "ce": 6.1894097328186035, "genuineness": 4.607763767242432, "identity": 0.7585487961769104, "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/casual_not_formal/a8efbf50b855b1ff6f8ab1e3875b6127c587ffc4fbd84758006854ba30aa5f84.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "a8fe927dc717baac781a71bf98278aa7f01e80c52786b4538b617e58fcef901d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4d47606d44bd7fbe80a2d3cb50d03487175dce987ca528072f482869f556d11", "generation_seconds_batch": 23.341999776894227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197251796722412, "target_z": -0.6035156232600032, "ce": 5.9431867599487305, "genuineness": 4.381826400756836, "identity": 0.32519543170928955, "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/casual_not_formal/a8fe927dc717baac781a71bf98278aa7f01e80c52786b4538b617e58fcef901d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "a9341d714524cfcf72a1ede6de163d13a225e62818a5489e69b987e37e155ac9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ce190563cc600730d973f8a8f997a4bf78b7ef957d39b3314012635e1f9ffd6", "generation_seconds_batch": 24.057310626027174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5774586200714111, "target_z": -1.8281249739432568, "ce": 6.332377910614014, "genuineness": 4.143336772918701, "identity": 0.6618545055389404, "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/casual_not_formal/a9341d714524cfcf72a1ede6de163d13a225e62818a5489e69b987e37e155ac9.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "20879325f6d561eacf74e130c3a7febd4e361f577dfcd60a14a9999e4b981309", "generation_seconds_batch": 16.99237280804664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.295151710510254, "target_z": -1.160156250670707, "ce": 6.109982967376709, "genuineness": 4.356721878051758, "identity": 0.6669448614120483, "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/casual_not_formal/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "a960602b31ac4abc69567f99b5892666d3b0915ca74f7c6f3c0c1004aadab3c0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c58040e0f43a3a92f90d99544b30b25c0cda7bfc405fae983bb285b38701db8", "generation_seconds_batch": 24.543578927987255, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8968777656555176, "target_z": -0.10766605667582024, "ce": 6.1894097328186035, "genuineness": 4.482243061065674, "identity": 0.5303614139556885, "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/casual_not_formal/a960602b31ac4abc69567f99b5892666d3b0915ca74f7c6f3c0c1004aadab3c0.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": "a977415a5554162ecc1cdc33e15e3777b7655ab01e7794de2f05aae1577a91d6", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "ce37663e3b010e7ca3a9b28c4115ba3b9264fdfafaf581ac3491ef49d9167c8a", "generation_seconds_batch": 15.7392909869086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.660761833190918, "target_z": -0.28710938483891113, "ce": 6.086154937744141, "genuineness": 4.507347106933594, "identity": 0.5640326142311096, "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/casual_not_formal/a977415a5554162ecc1cdc33e15e3777b7655ab01e7794de2f05aae1577a91d6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "a9b08c2324ad6eae5c2dc391a9cf57a4e6506f23220d87849d84c1a59bea27d9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e68ab6ce4b8a70ae0c2cb650e2bc7f73425c7de27f5fa17e21c6ab363b63926c", "generation_seconds_batch": 23.00830056006089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3941190242767334, "target_z": -0.49121095020182265, "ce": 6.229123115539551, "genuineness": 4.557555198669434, "identity": 0.5723854303359985, "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/casual_not_formal/a9b08c2324ad6eae5c2dc391a9cf57a4e6506f23220d87849d84c1a59bea27d9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "a9e820126d91335ef00980f3d20f294dbf6b1e3f88babe1d54d28c1ee5814611", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db663c6cfa22ce69f363073c7744ae1f1f18aad2b1629e3083f9bf63bf70537b", "generation_seconds_batch": 25.666605120990425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7278897762298584, "target_z": -0.021545409748764253, "ce": 6.213237762451172, "genuineness": 4.337893962860107, "identity": 0.31890109181404114, "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/casual_not_formal/a9e820126d91335ef00980f3d20f294dbf6b1e3f88babe1d54d28c1ee5814611.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "aa125c084f4c4b48a34d7a91cf522ce552de44fc8ac9368eebbcd824deefbfc6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84063a800be89e63a518c1a118033c2ab057afbf8b59a6f60f0855c85d5957ca", "generation_seconds_batch": 23.709864328964613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.54986834526062, "target_z": -0.1972656414305322, "ce": 6.078212261199951, "genuineness": 4.570107460021973, "identity": 0.3457225561141968, "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/casual_not_formal/aa125c084f4c4b48a34d7a91cf522ce552de44fc8ac9368eebbcd824deefbfc6.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": "aa1279dbe9f87414c65ca0bed7473962e2b240c92cc9e129f2be628314cf97dc", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "9f23a5584bfcf5105f8dee2c31c9d2a329f1985e06cf3490e9c4b6564da41ff0", "generation_seconds_batch": 24.059360969928093, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.291947364807129, "target_z": -0.5585937948541634, "ce": 6.006728172302246, "genuineness": 4.645420074462891, "identity": 0.6230940818786621, "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": 0.120000000000001, "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/casual_not_formal/aa1279dbe9f87414c65ca0bed7473962e2b240c92cc9e129f2be628314cf97dc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "aa3ae9df6625e5998fbdea51fba99bb9a8fffc65a253aa77800dfe41f5406e28", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1efc0061e608af9d28ea8a3ab07c72dd012f1dd9f5589e383b868fa79abb4c53", "generation_seconds_batch": 18.95441730809398, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.045952796936035, "target_z": -1.5273437114543291, "ce": 6.332377910614014, "genuineness": 4.268857479095459, "identity": 0.36975589394569397, "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/casual_not_formal/aa3ae9df6625e5998fbdea51fba99bb9a8fffc65a253aa77800dfe41f5406e28.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "aa4450fcec52f01c9b8232432823dbe23533d306aa37f63f564b0e155fb893a7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "91d66b0b5fb2a5df5af40a08dacd0d7c2a6a239fae2ae4a51cda3e6e695fbb05", "generation_seconds_batch": 24.37862454901915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.876497507095337, "target_z": -1.3515624894831784, "ce": 6.117925643920898, "genuineness": 4.20609712600708, "identity": 0.7249828577041626, "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/casual_not_formal/aa4450fcec52f01c9b8232432823dbe23533d306aa37f63f564b0e155fb893a7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "aa57b6baa1530984b19c29b8f84294ad3eb077e444fde7fa8b44d4247de8d286", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e8baf763b05aa80b97c6c3f215689ebdf4f958cbab40448fe9406e15b9815edb", "generation_seconds_batch": 20.710037415032275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3716912269592285, "target_z": -0.47070304991997997, "ce": 6.181467056274414, "genuineness": 4.570107460021973, "identity": 0.6499415040016174, "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/casual_not_formal/aa57b6baa1530984b19c29b8f84294ad3eb077e444fde7fa8b44d4247de8d286.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "aa9671ab06bac81e4f1ebb9c9e83e10d1029cc178e9b30061448ab7aea65e454", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3cd324ccfb43301a2eee5bf5d8510fa4c0b47de275d50ff3bfb27948352990f9", "generation_seconds_batch": 16.046525139012374, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3150875568389893, "target_z": -1.265624991656976, "ce": 5.95112943649292, "genuineness": 4.281409740447998, "identity": 0.5875656604766846, "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/casual_not_formal/aa9671ab06bac81e4f1ebb9c9e83e10d1029cc178e9b30061448ab7aea65e454.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "aaae8d662761e2f6fbd537f85a70ea659087964ed3c62853646d123f72565e17", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "154171c49295ea3ee41af46bfd3c39f0eeab7ce1a3cd835a0beb79520e6c8740", "generation_seconds_batch": 22.547705956036225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.551114320755005, "target_z": -0.2949218570485711, "ce": 5.959072113037109, "genuineness": 4.570107460021973, "identity": 0.6498546600341797, "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/casual_not_formal/aaae8d662761e2f6fbd537f85a70ea659087964ed3c62853646d123f72565e17.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": "ab1321280f56e93fdd3bec7e529c1ab986556f62595da7d91468db1f1e55287c", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b3bb86136b100ac47b51bc945eb831dff76d27b6a415fb039968595833a9ecd", "generation_seconds_batch": 25.921348684001714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -0.9023436547563369, "ce": 6.054384231567383, "genuineness": 4.532451152801514, "identity": 0.4922373592853546, "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/casual_not_formal/ab1321280f56e93fdd3bec7e529c1ab986556f62595da7d91468db1f1e55287c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "ab4eab7c98bc9e35aeedcf1c49bede9a6ddb359763bfefa6dafa28868cd2d5af", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6e96587f2a150e9d542479f8f781f47e8a08ea7c221e68019c464caf30053e30", "generation_seconds_batch": 17.97157822700683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6440300941467285, "target_z": -0.8867188259516194, "ce": 6.229123115539551, "genuineness": 4.595211505889893, "identity": 0.5269840955734253, "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/casual_not_formal/ab4eab7c98bc9e35aeedcf1c49bede9a6ddb359763bfefa6dafa28868cd2d5af.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "ed85e23d78e4e7b596412a1953093168aaa67d3c55641b31631d3308b19d3d45", "generation_seconds_batch": 15.827717834035866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9081811904907227, "target_z": -0.6074218271332552, "ce": 6.467403411865234, "genuineness": 4.394378662109375, "identity": 0.2705170512199402, "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/casual_not_formal/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "aba7ea8fa3e3f8d61bfdf63bfbf40fcb7e066433a8266fe5043000e55d208929", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2f4a482111c8599eb261d419aeac24075f2493836c4a8a24355b1576b78ab6cc", "generation_seconds_batch": 18.134781057946384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2994234561920166, "target_z": -0.342773385597006, "ce": 6.459460258483887, "genuineness": 4.645420074462891, "identity": 0.6986868381500244, "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/casual_not_formal/aba7ea8fa3e3f8d61bfdf63bfbf40fcb7e066433a8266fe5043000e55d208929.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "abe499469ef0149e05839c121eae67bb1cc72ba7f341ebc0c16cf5e8708ebe56", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f490335beb28d17e276b7eec6609a34e31119cb06714102ba669497fce9c21a1", "generation_seconds_batch": 15.98994576104451, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2495837211608887, "target_z": -0.5791015207352528, "ce": 6.24500846862793, "genuineness": 4.482243061065674, "identity": 0.5967628955841064, "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/casual_not_formal/abe499469ef0149e05839c121eae67bb1cc72ba7f341ebc0c16cf5e8708ebe56.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "acaa5a27b2305d209b8aacc9d3befedff1eb96850ff477ac66dd22e1a4feab86", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b749474d3e94446c44b903bdcca7d24a2af430280d9ef2e109930a5ef72669ee", "generation_seconds_batch": 22.393519739969634, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.9531250734416686, "ce": 5.935244083404541, "genuineness": 4.193545341491699, "identity": 0.6124495267868042, "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/casual_not_formal/acaa5a27b2305d209b8aacc9d3befedff1eb96850ff477ac66dd22e1a4feab86.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": "ace97d0cde3df0c52bc47ec53f2fd8fa8cc59020260bc7fdc56e99093d10ea14", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "ae1a34cf15b3909106898bf4b7e32e5660e512b0def7289b3b7394baa2efbea0", "generation_seconds_batch": 16.766787564964034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.079416275024414, "target_z": 0.12316903564176145, "ce": 6.157639026641846, "genuineness": 4.243753433227539, "identity": 0.545311689376831, "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/casual_not_formal/ace97d0cde3df0c52bc47ec53f2fd8fa8cc59020260bc7fdc56e99093d10ea14.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "ad1e4e3260dd7b7cfa3f7ab3c18d38904789b45a2e400c2b496808a928f2d9f6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d4a236ddce6bb5a3dba4d5476454255175cd212901b9a180a118181e73e0146d", "generation_seconds_batch": 23.590502972947434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.768073081970215, "target_z": -0.15618900732972785, "ce": 6.229123115539551, "genuineness": 4.7081804275512695, "identity": 0.7316635847091675, "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/casual_not_formal/ad1e4e3260dd7b7cfa3f7ab3c18d38904789b45a2e400c2b496808a928f2d9f6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "ad475b3a3558480343e6be7317384573cf68e3239204544f7721da6e49ce1ebd", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "723a8018619f72fec2c43c3ad6f272e4e1c057294757411ea83a4a77969dac25", "generation_seconds_batch": 17.670204249909148, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.4183268547058105, "target_z": 0.3237305767777251, "ce": 6.149696350097656, "genuineness": 4.7332844734191895, "identity": 0.37059417366981506, "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/casual_not_formal/ad475b3a3558480343e6be7317384573cf68e3239204544f7721da6e49ce1ebd.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "dad8d3c46a601673a92401059622834624ac6e72d3db8b2c62b8915ff1f0bdc1", "generation_seconds_batch": 17.258046666975133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.813485622406006, "target_z": -0.8671873861430035, "ce": 6.125868320465088, "genuineness": 4.444586753845215, "identity": 0.7010059356689453, "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/casual_not_formal/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "b649ac760efbda3b53d99078ea3e57e5470edd5df7d05d801a4d45d5857cae32", "generation_seconds_batch": 17.52830210106913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6372663974761963, "target_z": -1.6015625465077215, "ce": 6.38003396987915, "genuineness": 4.030367851257324, "identity": 0.37159067392349243, "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/casual_not_formal/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "ae85b4aa0079ebc90127d288c9196b858d09cf8e0e86369a86c8edbaebc6cb6f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "15e6829a61fd8932e63c581916e52e390cabd724510d28c2da8a04b69ff3c408", "generation_seconds_batch": 22.872305673081428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.237123727798462, "target_z": -0.27587889245587577, "ce": 5.967014789581299, "genuineness": 4.432034492492676, "identity": 0.3272441625595093, "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/casual_not_formal/ae85b4aa0079ebc90127d288c9196b858d09cf8e0e86369a86c8edbaebc6cb6f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "aea5910e163caf1ae03ba1a8192d74a37ac59327ed36c98bba0b1f7b49bac2e2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ace0ce248e5976d642f61db3f9c358bd5f50ac6376905c0d7d23ba39a2598c7e", "generation_seconds_batch": 17.711237773066387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3866429328918457, "target_z": -0.4238281471765679, "ce": 5.6413655281066895, "genuineness": 4.444586753845215, "identity": 0.3796316385269165, "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/casual_not_formal/aea5910e163caf1ae03ba1a8192d74a37ac59327ed36c98bba0b1f7b49bac2e2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "aec0a343e16925b396636420bb6a68ba38c8ffba5c0d5b51d30781161da99782", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1139e637004fe85e1f593a78de71c097619dd4127308fcd517e07f148e9d5847", "generation_seconds_batch": 17.95938637095969, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3991029262542725, "target_z": -0.33886719659937237, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.5485783219337463, "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/casual_not_formal/aec0a343e16925b396636420bb6a68ba38c8ffba5c0d5b51d30781161da99782.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "af55bc21e17219537d6ae7f649a3c628780dfc60a79edacfafa6b53ae00d3dd6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d836fb0b9698baa99f764101a3668fbebd0d443888393caa1ff922b32bec0fcf", "generation_seconds_batch": 18.194291541003622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7568819522857666, "target_z": -1.671875086343582, "ce": 6.292664527893066, "genuineness": 4.143336772918701, "identity": 0.7510482668876648, "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/casual_not_formal/af55bc21e17219537d6ae7f649a3c628780dfc60a79edacfafa6b53ae00d3dd6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "af6cb295029ade554a9e3257e4be4f82c404053532887e28d6b112aebbb31495", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "567739afcb6307ab5b9a7ee5a69e3b584d5c877040b2a364220c5d981e962de1", "generation_seconds_batch": 14.772496189922094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.4921875524528456, "ce": 5.804190158843994, "genuineness": 3.898571252822876, "identity": 0.6137732863426208, "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/casual_not_formal/af6cb295029ade554a9e3257e4be4f82c404053532887e28d6b112aebbb31495.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "af8ec473d301e16f0998c54edaf75213ca8fe2729360557c590c3e22bd9014a9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1050eb9ee0bcbaa4408571a2fbc78e0ebf0f83dfcdf0712a61b07f4646244824", "generation_seconds_batch": 24.544297170010395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4845428466796875, "target_z": -1.1992187376423356, "ce": 5.982900142669678, "genuineness": 4.055472373962402, "identity": 0.6745132803916931, "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/casual_not_formal/af8ec473d301e16f0998c54edaf75213ca8fe2729360557c590c3e22bd9014a9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "afb5cd4cc4ca76a3817d236c44cc71ae329bc65bda6faca4362e9081b82de476", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0fc2cb6e5a2020f69ff015c434b567ced4b0043004a48995a5584b9246559b8a", "generation_seconds_batch": 18.817252121050842, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8433892726898193, "target_z": -0.7304688120380408, "ce": 6.1894097328186035, "genuineness": 4.049196243286133, "identity": 0.35551172494888306, "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/casual_not_formal/afb5cd4cc4ca76a3817d236c44cc71ae329bc65bda6faca4362e9081b82de476.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "afb772894510891b6e4160fcb401cd8306e69ce40f62e8aa13f6b3525ac7cb41", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "022c12786a1a9021f0d0a63fe306286e077c85bd03ab0997bbedf72b6ee96b53", "generation_seconds_batch": 17.931166169932112, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.127476215362549, "target_z": -0.6113281631935077, "ce": 6.229123115539551, "genuineness": 4.482243061065674, "identity": 0.7975893020629883, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude auge 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/casual_not_formal/afb772894510891b6e4160fcb401cd8306e69ce40f62e8aa13f6b3525ac7cb41.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "afc0bc8534ae9f50a31a9704052c7b60dc0d0addfa49727578d5a7627f28dc50", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "63bac1f0f5888f9c7bada2d6ffade96aab8bbf5e950dee02d66e74327b4b383f", "generation_seconds_batch": 25.37599006900564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.614126205444336, "target_z": -1.0625000544473617, "ce": 6.086154937744141, "genuineness": 4.582659721374512, "identity": 0.6055847406387329, "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/casual_not_formal/afc0bc8534ae9f50a31a9704052c7b60dc0d0addfa49727578d5a7627f28dc50.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "afe231780a9b90ecf663e6bcc242d22913116ec081ef3a379f665fc81ad9fe18", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac072bc456a6105e09bdaf8b8f9502fe06b1c13e63a453bd9064cb0b3e89e24d", "generation_seconds_batch": 15.98994576104451, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718790054321289, "target_z": -0.9746092884387159, "ce": 6.173524379730225, "genuineness": 4.482243061065674, "identity": 0.6022822260856628, "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/casual_not_formal/afe231780a9b90ecf663e6bcc242d22913116ec081ef3a379f665fc81ad9fe18.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "b068b1a60999bc83c24458e1349541a23518b6c01f3dad0445bc0c7f8f216fe2", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07c8e0c6ab1a5eec4df32b29ae24a312c7f4fc9ce5ea97a2cc0ff59ca2c384b5", "generation_seconds_batch": 18.240847948007286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8035175800323486, "target_z": -0.6621093465397524, "ce": 6.2767791748046875, "genuineness": 4.293961524963379, "identity": 0.3675837814807892, "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/casual_not_formal/b068b1a60999bc83c24458e1349541a23518b6c01f3dad0445bc0c7f8f216fe2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "b0a293ed3bfaa32faa30a9adeb2c0264ffdff2ddf33a36ec41dd5ee01e548288", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "672e3e0c52661e7955253c99fe3fb290aadf1764514a56b37a1e72bfc2845fea", "generation_seconds_batch": 21.978913510916755, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.7851563048891617, "ce": 6.133810997009277, "genuineness": 4.6956281661987305, "identity": 0.5793673992156982, "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/casual_not_formal/b0a293ed3bfaa32faa30a9adeb2c0264ffdff2ddf33a36ec41dd5ee01e548288.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "b0ef0fdc31823e10ab1eb781805be9321399d68d88bce1e1988d4a612bafeee0", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0cc6cce30c6cf1d6669429aece063093d394310eeeb70c3486b68b31903148b4", "generation_seconds_batch": 18.717890892992727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.086892127990723, "target_z": 0.08593752239146563, "ce": 6.38003396987915, "genuineness": 4.570107460021973, "identity": 0.4529336094856262, "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/casual_not_formal/b0ef0fdc31823e10ab1eb781805be9321399d68d88bce1e1988d4a612bafeee0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "b189d50e3170a32483cfc3c3650a8f6a75fabd749e98d5f2a9d30d6b05f74b73", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b8bb0b3a7ffc214261071424b3891014def9ca844b08264c614ed35870cc7be", "generation_seconds_batch": 18.420794353005476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2670276165008545, "target_z": -0.20874018995813745, "ce": 5.919358730316162, "genuineness": 4.017816066741943, "identity": 0.25827348232269287, "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/casual_not_formal/b189d50e3170a32483cfc3c3650a8f6a75fabd749e98d5f2a9d30d6b05f74b73.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "b1fc301dfe2c95a489b317a1afd5bec7f183561031cdd34a9bd3fbdcec2963e3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ea6699c0a87965b18c3c6e4b4cb87db6d4b15d3b339e28e65d87775a438bf62c", "generation_seconds_batch": 25.791150578996167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.313663005828857, "target_z": 0.22009275578295528, "ce": 6.109982967376709, "genuineness": 4.532451152801514, "identity": 0.45725640654563904, "hypothesis": "La porte étant encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand il peut briquer.", "wer": 0.2222222222222222, "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/casual_not_formal/b1fc301dfe2c95a489b317a1afd5bec7f183561031cdd34a9bd3fbdcec2963e3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "b3611079b82a9bba15bc090c0861f9617ec11842f6022d86f1bec6debefd7a0d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f45cbd1aad1fadad0f8992b65ffc4cf7b1d528a6f45aaa6a4ce2c1761e5be66a", "generation_seconds_batch": 18.945967592997476, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8166897296905518, "target_z": -1.62109382855281, "ce": 6.252951145172119, "genuineness": 4.268857479095459, "identity": 0.7378168702125549, "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/casual_not_formal/b3611079b82a9bba15bc090c0861f9617ec11842f6022d86f1bec6debefd7a0d.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": "b41dad120711f9fe8913efd258499f6905095b4d512e2a3ff091ac8d688f0dd7", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2f877c7e46a97d4133b38f5b85a6fb9abc0bc0d5e7ee3440f6648a58862928b", "generation_seconds_batch": 18.769476873916574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4314987659454346, "target_z": -0.49414065680418107, "ce": 6.348263263702393, "genuineness": 4.344170093536377, "identity": 0.6714539527893066, "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/casual_not_formal/b41dad120711f9fe8913efd258499f6905095b4d512e2a3ff091ac8d688f0dd7.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "6864c187e0e605033c755f8fe440d8dabc536503b1001a977d4d5157fbe0a56a", "generation_seconds_batch": 16.548972547985613, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.504478931427002, "target_z": -1.1445311453568274, "ce": 6.324435234069824, "genuineness": 4.356721878051758, "identity": 0.562415361404419, "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/casual_not_formal/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "b4a7bdfac816d502f31d01608290ecc2dc8587e55bf4c19ce065579de5f774f7", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b86709862b9fee315efd128ecab39bf0413d9b590ef63191bf8baaeb2d7af626", "generation_seconds_batch": 16.602567621972412, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.2265625575335737, "ce": 6.427689552307129, "genuineness": 4.457139015197754, "identity": 0.5755913257598877, "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/casual_not_formal/b4a7bdfac816d502f31d01608290ecc2dc8587e55bf4c19ce065579de5f774f7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "b4c712e8383965ade80ddf3e675df762e4f6ff11600b0821e9d33d988bd322ea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac3965a89a9534066e65d2a07b155452bbb91ec5987d3c9103ed93a266460756", "generation_seconds_batch": 17.02038426406216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0726523399353027, "target_z": -0.6015624846553663, "ce": 6.205295085906982, "genuineness": 4.3881025314331055, "identity": 0.4873107373714447, "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/casual_not_formal/b4c712e8383965ade80ddf3e675df762e4f6ff11600b0821e9d33d988bd322ea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "b51c35425f2231dac900fca2b85bde841312f3618aef65fa4753bbd0db10e503", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e797f40d2dd548a5da1cd5bdaf907db25a8d9e43c5a2e50ea943999337a1a9bf", "generation_seconds_batch": 18.3308374020271, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.926692485809326, "target_z": 0.6738281092687883, "ce": 6.300607204437256, "genuineness": 4.444586753845215, "identity": 0.33554142713546753, "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/casual_not_formal/b51c35425f2231dac900fca2b85bde841312f3618aef65fa4753bbd0db10e503.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "b55e733c17775907590925bde686f315314034398a15004968feeb3dde8e1c2e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b17b225cf3bb7d7f6db2cf4e9baded87311da7982f5efeabd2d5e05ecf49b63f", "generation_seconds_batch": 26.885386556969024, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8420541286468506, "target_z": -0.04077147255017897, "ce": 6.1020402908325195, "genuineness": 4.231201171875, "identity": 0.7133186459541321, "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/casual_not_formal/b55e733c17775907590925bde686f315314034398a15004968feeb3dde8e1c2e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "b5a8f74bafe5692e4904b0d85108009c56148932d12c323fcbd678dd7fcc6c55", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c64f7555acdd5c72786f6657adeadfba87660b76d1affb2000fa220c8b622ce2", "generation_seconds_batch": 26.50834251602646, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0527164936065674, "target_z": -0.6328124705785121, "ce": 6.197352409362793, "genuineness": 3.848362922668457, "identity": 0.37914329767227173, "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/casual_not_formal/b5a8f74bafe5692e4904b0d85108009c56148932d12c323fcbd678dd7fcc6c55.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "b5af7e556102affcafdaa3d5aa1ab96d2be33f8c098e1da7f2ab541071981a60", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3b6e890c8ea8522fe3a8b98846ddb4b31643564b15dc001ac887791a4bdfbba4", "generation_seconds_batch": 22.252327944966964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.48881459236145, "target_z": -0.2910156429638353, "ce": 6.022613525390625, "genuineness": 4.482243061065674, "identity": 0.4550544023513794, "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/casual_not_formal/b5af7e556102affcafdaa3d5aa1ab96d2be33f8c098e1da7f2ab541071981a60.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "b5b93c5dba85dfc556f4578390032553aecb975e6e9584d420610fcd64e5f228", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "01ea5a2533f48b092bc0edcb441fa8bd74df1f30e35f789e47ccec8db2a43532", "generation_seconds_batch": 22.38128261792008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7261765003204346, "target_z": -0.06884770319742323, "ce": 6.086154937744141, "genuineness": 4.319066047668457, "identity": 0.25376421213150024, "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/casual_not_formal/b5b93c5dba85dfc556f4578390032553aecb975e6e9584d420610fcd64e5f228.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "b5c7fb78951548de00b85bc622f6d458a97604268e5223d8de8ff726fe9603ec", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f6549bd9ba06814d5a8d6e4cecd3da86a583b28cc5f9e7c89634998eebb58d31", "generation_seconds_batch": 25.191665918100625, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8900249004364014, "target_z": 0.30371093969740726, "ce": 5.895530700683594, "genuineness": 4.149612903594971, "identity": 0.5676688551902771, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/b5c7fb78951548de00b85bc622f6d458a97604268e5223d8de8ff726fe9603ec.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "b5e4559a5c1005af78591173494393d0c13fed75be5336c0d20f127b245b86f1", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bb1ec027be4533b06c11919dca24b4bde64480e748dd027c9c6526b38b8bfd7", "generation_seconds_batch": 19.03715557209216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.619110345840454, "target_z": -0.904296858736544, "ce": 6.205295085906982, "genuineness": 4.557555198669434, "identity": 0.5679447650909424, "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/casual_not_formal/b5e4559a5c1005af78591173494393d0c13fed75be5336c0d20f127b245b86f1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "b6408d3aa1bec60afe1d85cb222236542f81197b57a34a5a021ff6b81bc9ef63", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33ccb5e453816abea4395e9f770e6ffe0ae2c5c2672eacb14f6aae0214240ab9", "generation_seconds_batch": 19.280610741931014, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.9687500032462896, "ce": 6.24500846862793, "genuineness": 4.293961524963379, "identity": 0.638361930847168, "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/casual_not_formal/b6408d3aa1bec60afe1d85cb222236542f81197b57a34a5a021ff6b81bc9ef63.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "b682200d2ad210fe1d55d9a11734329dd3bfa2dc4c7331e3f578e703b7b36a40", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bb987c0c796bf1af7dd58f06317950a10dfdcd19e7acf2784fe562e48a8a2f5a", "generation_seconds_batch": 25.791150578996167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.982940673828125, "target_z": -0.8203125060395701, "ce": 6.133810997009277, "genuineness": 4.519899368286133, "identity": 0.7187891006469727, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/b682200d2ad210fe1d55d9a11734329dd3bfa2dc4c7331e3f578e703b7b36a40.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "b6dfea3e7b9f5377a94485df59869a445d496d95a17e3503a7e014309fa3603a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "53bd40da1bb4306358349f534368b9670030319146eaf46e9f61e888ad7bbb32", "generation_seconds_batch": 25.37599006900564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.723773956298828, "target_z": -1.0234374518611307, "ce": 6.1894097328186035, "genuineness": 4.469690799713135, "identity": 0.6801202893257141, "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/casual_not_formal/b6dfea3e7b9f5377a94485df59869a445d496d95a17e3503a7e014309fa3603a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "b6e0f7478381775ed88505902835a080eb0a3b65b0a18b3945b4e831a95e5e88", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2887a749bc97986123b9bca95105c7e08f6861ea9ae4c0a651d40b817137e728", "generation_seconds_batch": 25.856807679054327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.551114320755005, "target_z": -0.0688476933696168, "ce": 6.086154937744141, "genuineness": 4.243753433227539, "identity": 0.7695763111114502, "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/casual_not_formal/b6e0f7478381775ed88505902835a080eb0a3b65b0a18b3945b4e831a95e5e88.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "b768628add579c83db34ce7b768c37a845eb8ee0bdbddca166a4a49fff830764", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d8e169f9d7f70f79f4c481812fcd54b5f4c7afcf1da3e6d095979f57ef36a515", "generation_seconds_batch": 25.12612471205648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7387256622314453, "target_z": -1.0214845178654945, "ce": 6.475345611572266, "genuineness": 4.2751336097717285, "identity": 0.5827082395553589, "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/casual_not_formal/b768628add579c83db34ce7b768c37a845eb8ee0bdbddca166a4a49fff830764.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": "b77a43ad2cf5394cb8c6c7ffa84a0f7c4367f90bc1e518248c812ae271efa614", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "cbb4ce064f88198d3462f9d50933ed7f7475983c481a61946cf8fb6cadcd2ab5", "generation_seconds_batch": 18.035754844895564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.670729875564575, "target_z": -0.314453138571821, "ce": 5.998785495758057, "genuineness": 4.645420074462891, "identity": 0.6456934213638306, "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/casual_not_formal/b77a43ad2cf5394cb8c6c7ffa84a0f7c4367f90bc1e518248c812ae271efa614.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "b78f9c6f1ea4be7f4b35684026b12e94a4a518431b9ef00c7aac3334e8b6fc8d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "817201a4433b2bc2eb8b7d4fefa44baf500f1efc5861811462286385d1ced02a", "generation_seconds_batch": 22.654940251028165, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7274227142333984, "target_z": -0.3371581609932842, "ce": 6.205295085906982, "genuineness": 4.607763767242432, "identity": 0.8653092384338379, "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/casual_not_formal/b78f9c6f1ea4be7f4b35684026b12e94a4a518431b9ef00c7aac3334e8b6fc8d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "b7a82d34f81427d53240a49951d4712d3dda1bfe960eb51ec3af2108f57fdc32", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eebb62877de660bbf97294cd5ba5655ab61c30a3df743063349715d7ef5b82c6", "generation_seconds_batch": 18.664490289986134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1075401306152344, "target_z": -0.6933593826371025, "ce": 5.863759994506836, "genuineness": 4.193545341491699, "identity": 0.6867799758911133, "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/casual_not_formal/b7a82d34f81427d53240a49951d4712d3dda1bfe960eb51ec3af2108f57fdc32.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "b7b97f20a0799f61cc3b8315a084cb8df8a7863bbb85f5a1f293aeb7a94fb233", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f3be9525eeec521f3bdef9776b7aa2987b6d33ed6df3b2cfb15ab5edaf69c2a6", "generation_seconds_batch": 22.362256914027967, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7028141021728516, "target_z": -0.2843017858147383, "ce": 6.078212261199951, "genuineness": 4.457139015197754, "identity": 0.7496950626373291, "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/casual_not_formal/b7b97f20a0799f61cc3b8315a084cb8df8a7863bbb85f5a1f293aeb7a94fb233.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "b7ebacfddaf23b1128a0f936454c7d6480ddecda9da212a35402ed68e6a8a2a9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1112bbc9b24723909026d6e6150603529abfbea0facec2667150351c1e121ba2", "generation_seconds_batch": 17.711237773066387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6785173416137695, "target_z": -0.0699463015219111, "ce": 6.1894097328186035, "genuineness": 4.300237655639648, "identity": 0.6611689925193787, "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/casual_not_formal/b7ebacfddaf23b1128a0f936454c7d6480ddecda9da212a35402ed68e6a8a2a9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "b8077266e7a0c195c92705bc5490828504524cbaf1ed90014ec9e366217d84ef", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a81005c87d0ba56c6a60096aef835167e5e6e0907aa35093faf4d7bdfb8ad77f", "generation_seconds_batch": 25.828512109001167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246379375457764, "target_z": 0.20947265602489656, "ce": 6.348263263702393, "genuineness": 4.519899368286133, "identity": 0.21461495757102966, "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/casual_not_formal/b8077266e7a0c195c92705bc5490828504524cbaf1ed90014ec9e366217d84ef.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": "b825982dfac7d274776934e8343514cbe3028fd3afe9b521f07b38d9de5c2a16", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8043108acf86980af9ba5f87a8909aeea4f49f15234352c9f941f4875c6689d9", "generation_seconds_batch": 17.471166706061922, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.4345702655380702, "ce": 6.252951145172119, "genuineness": 4.3504462242126465, "identity": 0.4276518225669861, "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/casual_not_formal/b825982dfac7d274776934e8343514cbe3028fd3afe9b521f07b38d9de5c2a16.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "b92b40e5fdb685504d11ac6af56bb8fa29bc7d157353e702a1cae72b4b8a86ff", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "133206eabf06c1c3dc5e78e4e28a63ccf42fd5060e1f712ee119e6936b488c98", "generation_seconds_batch": 23.014348095981404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.868309259414673, "target_z": -0.9394531137788558, "ce": 6.125868320465088, "genuineness": 4.21237325668335, "identity": 0.7423262596130371, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/b92b40e5fdb685504d11ac6af56bb8fa29bc7d157353e702a1cae72b4b8a86ff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "b961c5f23381bbaf18c403bcb882edbfdc2b56279e7d3be62a61628ef9aa7d04", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "608e062d2bc77689f1556b5f00a961e9cd5e44c8ffb3f28cb88ff97c9747896f", "generation_seconds_batch": 22.858587356982753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.617151975631714, "target_z": -0.27783208770988377, "ce": 6.213237762451172, "genuineness": 4.432034492492676, "identity": 0.7479982972145081, "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/casual_not_formal/b961c5f23381bbaf18c403bcb882edbfdc2b56279e7d3be62a61628ef9aa7d04.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": "b9d764c7f04bec8f80f1673248d87fca9a1ffbbb6e88447b17d344fc07cb25f1", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "169d0d5018e9d1d8dc2fa4ca693551ddbce8c1f84ac7e8333d45f65c689f5216", "generation_seconds_batch": 18.653824200038798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7387256622314453, "target_z": -0.8945313129720698, "ce": 6.395919322967529, "genuineness": 4.168440818786621, "identity": 0.2808969020843506, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gates. I will meet you on the other side.", "wer": 0.041666666666666664, "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/casual_not_formal/b9d764c7f04bec8f80f1673248d87fca9a1ffbbb6e88447b17d344fc07cb25f1.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "b9061774dfb4392b5c20e457069a735261557a5d3d1c99ea4673c0bdd5b1564f", "generation_seconds_batch": 17.74653109605424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3991029262542725, "target_z": -0.5751953239084309, "ce": 6.24500846862793, "genuineness": 4.595211505889893, "identity": 0.7170396447181702, "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/casual_not_formal/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "bac7c738838ba0464a299ff245c5180bf81af041af68cd2849e186f64d49c302", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f48d950c38d179df42c8075b419e8357b7a57348f39d044f5b32e8a1de65f10", "generation_seconds_batch": 18.1723309190711, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972972869873047, "target_z": -0.5478515381068166, "ce": 6.23706579208374, "genuineness": 4.369274139404297, "identity": 0.3856533169746399, "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/casual_not_formal/bac7c738838ba0464a299ff245c5180bf81af041af68cd2849e186f64d49c302.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "bad7dba3ee15da1fd348c0dd179475a0a6307daa2a2be8db5188917db79845ac", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b9778943d9b8264c8416dea1c3bc0d3198ef65ab9a878b62346aa05ad56bdf1", "generation_seconds_batch": 22.52777690498624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9946885108947754, "target_z": 0.23242190099111562, "ce": 6.062326908111572, "genuineness": 4.231201171875, "identity": 0.7472386360168457, "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/casual_not_formal/bad7dba3ee15da1fd348c0dd179475a0a6307daa2a2be8db5188917db79845ac.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "bb0182957bb4461362fe0791fab677070a96e2b7b97406cd13859e4e0d3a5f2e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bdd9631c57e4a200edf8def52a1b5b6081a7ef095622810d4f6ab9eeffa7def8", "generation_seconds_batch": 17.827782398089767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.564286708831787, "target_z": -1.0624999097330172, "ce": 6.395919322967529, "genuineness": 4.432034492492676, "identity": 0.6262261271476746, "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/casual_not_formal/bb0182957bb4461362fe0791fab677070a96e2b7b97406cd13859e4e0d3a5f2e.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": "bb8c432ef42d48dfff2f63f90786ea23096fd27afb12438deae2d98abf45c934", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05bccfec1cfcb56fe1ff18fcae143d04311b2ec50e655faf83cce4baf4d42942", "generation_seconds_batch": 18.86433734698221, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4501888751983643, "target_z": -0.43798828112754234, "ce": 6.1894097328186035, "genuineness": 4.005263805389404, "identity": 0.21110200881958008, "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/casual_not_formal/bb8c432ef42d48dfff2f63f90786ea23096fd27afb12438deae2d98abf45c934.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "bb9c20cefdc1cf1587ec4e2642df7837e3d524c0601995000cda54d4d141ee1f", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdbddc73e5854863bcec271dc41f025aef67820f691c66b326540ae3787eb023", "generation_seconds_batch": 15.898526731994934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.08262038230896, "target_z": -0.6894530778287751, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.8932694792747498, "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/casual_not_formal/bb9c20cefdc1cf1587ec4e2642df7837e3d524c0601995000cda54d4d141ee1f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "bbd5e563d530dbe7edc1fb2b1c9bfa6392b44394a2ef4ff91388ea668720264a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "232d536b03b87431ebd6e16f1fd1595995e1feceb414bb65809fb279bd2585ce", "generation_seconds_batch": 18.3308374020271, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.873203992843628, "target_z": -0.09936524569378852, "ce": 6.181467056274414, "genuineness": 4.231201171875, "identity": 0.4198722839355469, "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/casual_not_formal/bbd5e563d530dbe7edc1fb2b1c9bfa6392b44394a2ef4ff91388ea668720264a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "bbde5f3b0b9bb048c4927278f807e355471ea09d15994587aef2ea130350e22a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fead87cb6182b1ef636b143232ae18377906e63abb99f628d4042a11c0185bc", "generation_seconds_batch": 21.740796232013963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.388422966003418, "target_z": 0.27783207973402885, "ce": 6.252951145172119, "genuineness": 4.6956281661987305, "identity": 0.42521560192108154, "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/casual_not_formal/bbde5f3b0b9bb048c4927278f807e355471ea09d15994587aef2ea130350e22a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "bc3db7f772a1657d79307b069f089c3067817620dd439b52c7411b3a06d6e610", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf5481bae68adc604159242a698aa97a84bba91f25612817355461840a66e18e", "generation_seconds_batch": 24.989573237020522, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0178287029266357, "target_z": -0.7675781187779462, "ce": 6.141753673553467, "genuineness": 4.519899368286133, "identity": 0.6946609616279602, "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/casual_not_formal/bc3db7f772a1657d79307b069f089c3067817620dd439b52c7411b3a06d6e610.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": "bc3dda62feb0f78170f96a16a3f23ecdaf194a362eea21d41c884d66eeb6d553", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 128, "seconds_generated": 10.24, "mp3_sha256": "7932d3e63974b57e324a3b7e45dab68ad9f324145e96222d951d701ee53c81fc", "generation_seconds_batch": 17.0212977839401, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9530367851257324, "target_z": -0.9648438749260704, "ce": 6.308549880981445, "genuineness": 4.300237655639648, "identity": 0.6194896697998047, "hypothesis": "Las llaves están en la mesa de la cocina, miren el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/bc3dda62feb0f78170f96a16a3f23ecdaf194a362eea21d41c884d66eeb6d553.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "bd6405531282343073a4c0ed7d86f9e61ba1b598457498e52fb92167ddee70de", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d2a60a736c66d4579b1410b76ac85062a97833a6ea1be377cad429966b3ca08c", "generation_seconds_batch": 17.349169188993983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.63210391998291, "target_z": -0.01806641804460815, "ce": 5.764476776123047, "genuineness": 4.3881025314331055, "identity": 0.3626369833946228, "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/casual_not_formal/bd6405531282343073a4c0ed7d86f9e61ba1b598457498e52fb92167ddee70de.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": "bdbc2640e479e18919f3dce0edd8d122c3ebb0f81c1a23434f326b25bc3318e8", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "88dfd52e5466c694c2f126cdaeade1518d0c5cf21ff6bcaf439929a71fe6d42f", "generation_seconds_batch": 18.68792234605644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.483830690383911, "target_z": -0.34570311239221946, "ce": 6.23706579208374, "genuineness": 4.570107460021973, "identity": 0.5627398490905762, "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/casual_not_formal/bdbc2640e479e18919f3dce0edd8d122c3ebb0f81c1a23434f326b25bc3318e8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "bdf14c384eea5ef1ba5010a3e8dd68913c54827210fd8876f7c4dbfd393fdfc7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5dcb14e9fb1d07949b17f2047d07dc80baf5dc5b48bdca8f6efae25a604a1a1f", "generation_seconds_batch": 21.522072001942433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1224923133850098, "target_z": -0.6464842930304213, "ce": 5.546053409576416, "genuineness": 3.760498285293579, "identity": 0.36998215317726135, "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/casual_not_formal/bdf14c384eea5ef1ba5010a3e8dd68913c54827210fd8876f7c4dbfd393fdfc7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "be174d292ba40cfc43fb1c2bcb4aabc0613f11e30edafe9a003edd962778b73f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf1dde31fb3d903a3ee5946466488e34b8c3fd425670e5363b7192e60e9067ab", "generation_seconds_batch": 26.607085130061023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.746112585067749, "target_z": -0.22924803276693154, "ce": 5.998785495758057, "genuineness": 4.532451152801514, "identity": 0.673161506652832, "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/casual_not_formal/be174d292ba40cfc43fb1c2bcb4aabc0613f11e30edafe9a003edd962778b73f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "bf71a2e0110aa97ed28c4c7539285c577a0a6ab15bca08f1192c01076db5a51f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aa3d929f779c4d57805c11c5334fec7983f7901fa5b2f35006f71a52313200ec", "generation_seconds_batch": 22.547705956036225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.506258487701416, "target_z": -0.2421875116749528, "ce": 5.919358730316162, "genuineness": 4.65797233581543, "identity": 0.837257981300354, "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/casual_not_formal/bf71a2e0110aa97ed28c4c7539285c577a0a6ab15bca08f1192c01076db5a51f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "bf8c315b90659c88a93cb281fa8c92b88126080c904ac2823008cb3a13a9034b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4feed42444624c578b37b7693bbfdb3b1a61251de251500dc9061616ce68c291", "generation_seconds_batch": 23.67371809692122, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.702502727508545, "target_z": -0.23950195700557855, "ce": 5.935244083404541, "genuineness": 4.545003414154053, "identity": 0.6717926263809204, "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/casual_not_formal/bf8c315b90659c88a93cb281fa8c92b88126080c904ac2823008cb3a13a9034b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "bfaf62b16e1bbfde71ba6df0636f37708b44755a8cbe391caab4986e3154409e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06b09ff744278fbcd06d9307ec8742ad0292a905e244621374ea367caebca134", "generation_seconds_batch": 23.511907604988664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.288743019104004, "target_z": 0.2993163330294312, "ce": 6.149696350097656, "genuineness": 4.419482707977295, "identity": 0.43556535243988037, "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/casual_not_formal/bfaf62b16e1bbfde71ba6df0636f37708b44755a8cbe391caab4986e3154409e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "bfd0594a1a83de89031c4a6df5364676f34ebe4d0e53f41bee29f8799a0dfa3a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "feb3c380524dd4a2be475f32a2aa22d815c6e5a34c1d8ef98e3f1bbbe56442ae", "generation_seconds_batch": 22.73846703604795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8339550495147705, "target_z": -0.15380865009623174, "ce": 6.181467056274414, "genuineness": 4.570107460021973, "identity": 0.7186347246170044, "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/casual_not_formal/bfd0594a1a83de89031c4a6df5364676f34ebe4d0e53f41bee29f8799a0dfa3a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c01fd2072dabe81caa605ce3338b806dc6611cd7eb66e9e3c252818ad07ec9cc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab5b7ebf7fcdd173720c17746af94ffa73fcae8d9baeb34cb6040b6a0f397431", "generation_seconds_batch": 24.688993592979386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4845428466796875, "target_z": -1.1054686859194247, "ce": 6.252951145172119, "genuineness": 4.369274139404297, "identity": 0.5301372408866882, "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/casual_not_formal/c01fd2072dabe81caa605ce3338b806dc6611cd7eb66e9e3c252818ad07ec9cc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "c06507014833c1985f460c57ae8b6eeaee4ae3d967d315d58d93f5b2057abe35", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdee85b7a26358d9cab507239c4e4429c8a0ca6b3c6a0d43fde2a1c64903786a", "generation_seconds_batch": 18.240847948007286, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.295151710510254, "target_z": -1.1914062285037454, "ce": 6.213237762451172, "genuineness": 3.9425034523010254, "identity": 0.4998038709163666, "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/casual_not_formal/c06507014833c1985f460c57ae8b6eeaee4ae3d967d315d58d93f5b2057abe35.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": "c08cc1f828576a1594c5923eac57fc122d6182e999074d50e51fd446602af354", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7fe30368b9d38c50e28342a5b1a4db8e70f64baf38399cbd8a933d94e59e7a43", "generation_seconds_batch": 18.43345167802181, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.523702383041382, "target_z": -0.3486328592182178, "ce": 6.213237762451172, "genuineness": 4.494794845581055, "identity": 0.7268753051757812, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis moi à quand nous pourrons parler.", "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/casual_not_formal/c08cc1f828576a1594c5923eac57fc122d6182e999074d50e51fd446602af354.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "c126cf88789a524c3e4961866a81d99d2921c6d32247d3b87eb5e6bf29dcce97", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e366ed7fb9104849494184cb90bb50cf8f5315be7e0a7cb96f9d3c95b5d7cf7e", "generation_seconds_batch": 24.896861342946067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.813485622406006, "target_z": -0.8749999692691102, "ce": 6.086154937744141, "genuineness": 4.432034492492676, "identity": 0.6539734601974487, "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/casual_not_formal/c126cf88789a524c3e4961866a81d99d2921c6d32247d3b87eb5e6bf29dcce97.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "c127f74a343465f8bb9ba6e79d0bc38346cde24a816b68536dd1aca6b8d018e4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1aded2ba73f6fd56b09a4ca6e058b36aeda4c57c0e13f8ace042a638837442b9", "generation_seconds_batch": 21.579793446930125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5635743141174316, "target_z": -0.23632812329641872, "ce": 6.0146708488464355, "genuineness": 4.406930446624756, "identity": 0.6471174955368042, "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/casual_not_formal/c127f74a343465f8bb9ba6e79d0bc38346cde24a816b68536dd1aca6b8d018e4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "c18701eb925546583912df7dfc96e33571a34276fddcb318d477de826166b322", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "440c5de9542d2bfedf25e0e6cf048f0cd13353e96f4c6fa4aea06e4ca0283a03", "generation_seconds_batch": 23.1512352040736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8794338703155518, "target_z": -0.08715823684125747, "ce": 6.260893821716309, "genuineness": 4.394378662109375, "identity": 0.4123683571815491, "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/casual_not_formal/c18701eb925546583912df7dfc96e33571a34276fddcb318d477de826166b322.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "c1b2d11c01d48078e86761b9ae4d14711b8176a3e6b13c2ae8399d7641f86b5e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3a7209ab51c3f389741a2419716c25cd7020ae271a041c366f920ac28542b13a", "generation_seconds_batch": 26.39916119107511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7143397331237793, "target_z": -0.2993163351013828, "ce": 6.348263263702393, "genuineness": 4.2751336097717285, "identity": 0.20698469877243042, "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/casual_not_formal/c1b2d11c01d48078e86761b9ae4d14711b8176a3e6b13c2ae8399d7641f86b5e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c1d5c795374c3c88a6c6bac76e4f1067aa6228210d18a796dec7af3b3bb49fff", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1925e1ac8d4ac2d71425d4f30031f504bd6cc4be9a7e21b50e2cf895f9a977fb", "generation_seconds_batch": 22.865506227943115, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1556003093719482, "target_z": -1.2500000769888928, "ce": 6.356205940246582, "genuineness": 4.432034492492676, "identity": 0.44917213916778564, "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/casual_not_formal/c1d5c795374c3c88a6c6bac76e4f1067aa6228210d18a796dec7af3b3bb49fff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "c1de89b865acdb69d025923c624e333865aaf1a5a08c4f9bdc8f72f5418236b5", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bcd3150851467056bbcf855e9882154807290164637c13faa9f517e1915b3e3c", "generation_seconds_batch": 16.112435367074795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.8007812985038366, "ce": 6.149696350097656, "genuineness": 4.105680465698242, "identity": 0.2720053493976593, "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/casual_not_formal/c1de89b865acdb69d025923c624e333865aaf1a5a08c4f9bdc8f72f5418236b5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "c1ffab12f03667994fcd55a5637a8386c54afbc46dc3552cdfd6a6e4d0f0823a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7bf7c9bf3c1e666fc21c939cce39bb7b793ec2ff274fbba28698f29f205c3ca8", "generation_seconds_batch": 17.148230073973536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8725810050964355, "target_z": 0.044433604090140515, "ce": 6.1020402908325195, "genuineness": 4.457139015197754, "identity": 0.7020943760871887, "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/casual_not_formal/c1ffab12f03667994fcd55a5637a8386c54afbc46dc3552cdfd6a6e4d0f0823a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "c23ba245ee6fb3efc975e2098f7497a7f2e4b265e8e948e40a0fdf5fab4ef370", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b737f75e78974b0132cdb99b13af2ad1924fce73769b19620407c1455b58defb", "generation_seconds_batch": 15.898526731994934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1997437477111816, "target_z": -0.5615234577420808, "ce": 6.070269584655762, "genuineness": 4.7332844734191895, "identity": 0.6794748902320862, "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/casual_not_formal/c23ba245ee6fb3efc975e2098f7497a7f2e4b265e8e948e40a0fdf5fab4ef370.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "c30467ea3c91c238a43d2eb295b68877f812d70f669359f260ee8ade6623b683", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e992fdfe1996293f6ae06e7edf824fc39767a2c2de5739d7f11388285c65bfa5", "generation_seconds_batch": 18.265823861933313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8035175800323486, "target_z": -0.9570312561756302, "ce": 6.133810997009277, "genuineness": 4.293961524963379, "identity": 0.5379778146743774, "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/casual_not_formal/c30467ea3c91c238a43d2eb295b68877f812d70f669359f260ee8ade6623b683.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "c33db07ecaa7d6fae7caeef02d65d4e700514c99c7fcda9e1b18e8128439207f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "05ef5070753b45a896a8eb105699ee781626742a800e9e340506e341d990192f", "generation_seconds_batch": 26.02787971496582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.456418752670288, "target_z": -0.449218785205739, "ce": 6.1894097328186035, "genuineness": 4.645420074462891, "identity": 0.49812909960746765, "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/casual_not_formal/c33db07ecaa7d6fae7caeef02d65d4e700514c99c7fcda9e1b18e8128439207f.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "c0b512c9dca7d402fa332dc293774ce6405c841e8ecc9f38f36c19b4bb9007ac", "generation_seconds_batch": 15.968530033016577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.624094247817993, "target_z": -1.1289063100275185, "ce": 6.197352409362793, "genuineness": 4.331617832183838, "identity": 0.7114896774291992, "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/casual_not_formal/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c3ca43ffc77d6fc235f6098867fc86d5c4c8b854607ba755a409cccfe8b48d77", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f648d8294b5bec975935d148a762192123a7691a2bc891575ae9ac8cdd7ef9fd", "generation_seconds_batch": 17.02038426406216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.619140553651291, "ce": 6.260893821716309, "genuineness": 4.632867813110352, "identity": 0.6765189170837402, "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/casual_not_formal/c3ca43ffc77d6fc235f6098867fc86d5c4c8b854607ba755a409cccfe8b48d77.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "c3e727a42420eaf8dd4d6d5d132ba31be18d6b3e3863dd2a29199fe41ce35e02", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b15e89bb7bae77607cd2de006ac57e242e3cf65d258c5399a50b1d9e1ff50a22", "generation_seconds_batch": 18.01441107899882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.182299852371216, "target_z": -0.6562500272236809, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.6898162364959717, "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/casual_not_formal/c3e727a42420eaf8dd4d6d5d132ba31be18d6b3e3863dd2a29199fe41ce35e02.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "c45ec3d6e6626f228ec149ea23ba4ed577f93b1805ce06888d9770d68ba2f956", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18f4c2d1dd4268e6e957c870ecfa7a832fd249e648d94ee91fa5e43a55a97f97", "generation_seconds_batch": 27.408408271032386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.490060567855835, "target_z": -0.3676757852670421, "ce": 6.038498878479004, "genuineness": 4.331617832183838, "identity": 0.5721511840820312, "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/casual_not_formal/c45ec3d6e6626f228ec149ea23ba4ed577f93b1805ce06888d9770d68ba2f956.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "626ce9da07aac9cc9955daaca53d70690dd795b0b1ad6eb9ff0d2d30daffd7ca", "generation_seconds_batch": 16.119336390984245, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6938700675964355, "target_z": -0.863281270871967, "ce": 6.324435234069824, "genuineness": 4.557555198669434, "identity": 0.7873736619949341, "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/casual_not_formal/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c639c543bf83e751d00c8110249ac7eec578ae80b1d2b1a6231089e342f4cda4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c0227dc89fb2c074fa6b3d6ca698da6f2371513e22d255f22385d7a688194e27", "generation_seconds_batch": 22.59962709294632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192267894744873, "target_z": -0.5097656093136087, "ce": 6.1894097328186035, "genuineness": 4.406930446624756, "identity": 0.4382324516773224, "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/casual_not_formal/c639c543bf83e751d00c8110249ac7eec578ae80b1d2b1a6231089e342f4cda4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c649112fb936a23d3873156803b8616c9f79f29375b7e305bd88fc85242c6b58", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ebe0992ee99b36405391ab502237067fa362b6845fd288fa504479c7fbfd4314", "generation_seconds_batch": 25.33540375402663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1157283782958984, "target_z": -1.371093747257944, "ce": 6.09409761428833, "genuineness": 4.256305694580078, "identity": 0.5111600160598755, "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/casual_not_formal/c649112fb936a23d3873156803b8616c9f79f29375b7e305bd88fc85242c6b58.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "c65a2ebce095b829576b42672a6246307ab02368990314d0ce3a746f69ab3ca2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6ce900001437e6f96790c9bde18d56f5940e31b78cdcb40b2eb061fee938741c", "generation_seconds_batch": 22.089328057016246, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4813387393951416, "target_z": -0.5273437678257702, "ce": 6.078212261199951, "genuineness": 4.545003414154053, "identity": 0.7295937538146973, "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/casual_not_formal/c65a2ebce095b829576b42672a6246307ab02368990314d0ce3a746f69ab3ca2.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": "c674ec886098db1919855e31692e18d6b7c6c8a8a3d7c03390bb8b962c9dbdc5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Someone moved the little blue chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "3b093fa95bdf2a72036cc978f71fab74d017daea9f52289af1fa04ab53109173", "generation_seconds_batch": 18.93227247998584, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4975366592407227, "target_z": -0.4506835772088159, "ce": 6.284721851348877, "genuineness": 4.20609712600708, "identity": 0.4314010739326477, "hypothesis": "Someone moved the little blue chair. Used to stand here beside the tall bookshelf. We can put it back after lunch.", "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/casual_not_formal/c674ec886098db1919855e31692e18d6b7c6c8a8a3d7c03390bb8b962c9dbdc5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c697204fd8bdcf21c8c3ca9ffc4725bd805c42625a0e6633e798e553ad0b2a0c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "afaec783f6e1d73a91669e7cd15b2b82471cc92513f5d85e1aae975955df1f49", "generation_seconds_batch": 22.600910209934227, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8433892726898193, "target_z": -0.8398437935655727, "ce": 6.221180438995361, "genuineness": 4.607763767242432, "identity": 0.5476217269897461, "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/casual_not_formal/c697204fd8bdcf21c8c3ca9ffc4725bd805c42625a0e6633e798e553ad0b2a0c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "c6d7ef82cee80993a13455b384403006a1a8c125b20d051e38c6b5d11876c8d1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b1528ee5d7ab6062fac7ad4bb979cbcd0d53540d85d3232f025eeba366380cb7", "generation_seconds_batch": 23.331219194922596, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.713805913925171, "target_z": -0.9843750598866532, "ce": 6.268836498260498, "genuineness": 4.243753433227539, "identity": 0.5727766752243042, "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/casual_not_formal/c6d7ef82cee80993a13455b384403006a1a8c125b20d051e38c6b5d11876c8d1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "c6e83406cd74e791ac27db0755ea95d81cd79c9fea9ec0031b71f4c4861ff73e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a2d71d0fe2bcef4a087f29846b2f4e833c0503f6434e8621dfe9f7a26f831263", "generation_seconds_batch": 24.592695421888493, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5393667221069336, "target_z": -1.0761718034984211, "ce": 6.117925643920898, "genuineness": 4.287685871124268, "identity": 0.6611934304237366, "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/casual_not_formal/c6e83406cd74e791ac27db0755ea95d81cd79c9fea9ec0031b71f4c4861ff73e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "c6e9810652486ea1044476ee1320d3e1c92ccf2bb4bbbfb39f9a5a81a2148bdc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c08ec65e8677da7cd81e1ac42f1a0cda7cb7b9530f35351d666d92acf6580bd7", "generation_seconds_batch": 23.922581055085175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -0.9960937632428715, "ce": 6.197352409362793, "genuineness": 4.2751336097717285, "identity": 0.6836000680923462, "hypothesis": "Ik heb het kleine doosje gevonden, het stond naast het oude raam. We kunnen het even samen in kijken.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/c6e9810652486ea1044476ee1320d3e1c92ccf2bb4bbbfb39f9a5a81a2148bdc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "c6ea69ca24cbd5c1f97febc74cba8e851c635021f4e5c983802b24154d67a208", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c9b4b652cfb4d0232a21cbf17b157107c8d3f08739d3854c7e45de88a9565168", "generation_seconds_batch": 17.82586977595929, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9043538570404053, "target_z": -0.11840819359983328, "ce": 6.252951145172119, "genuineness": 4.68307638168335, "identity": 0.6366192102432251, "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/casual_not_formal/c6ea69ca24cbd5c1f97febc74cba8e851c635021f4e5c983802b24154d67a208.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "c7030ba843b7e394cbe3d1e45643852956e105525d604f2efcd2fffbe7466241", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7021faefcca45c07947b5d4d4117480c946e0ae010363acf09329bc15933a5b4", "generation_seconds_batch": 18.46596200799104, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524414539337158, "target_z": -1.105468906186802, "ce": 6.141753673553467, "genuineness": 4.2751336097717285, "identity": 0.8508026003837585, "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/casual_not_formal/c7030ba843b7e394cbe3d1e45643852956e105525d604f2efcd2fffbe7466241.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "c749900cb983b929f1b7493352b934f9d80f7c2017c1d6a2e0dd6f988d7b9182", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ffe718c04b4767c2c1d8d7a39921440245c33c4c7f4c6733bad4137490951b91", "generation_seconds_batch": 18.56603705091402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7458009719848633, "target_z": -0.11413575859722327, "ce": 5.967014789581299, "genuineness": 4.582659721374512, "identity": 0.7817810773849487, "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/casual_not_formal/c749900cb983b929f1b7493352b934f9d80f7c2017c1d6a2e0dd6f988d7b9182.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "c771a65d208fde4288bf16dacbf4e9b17d72fe9109eedb6a30edde865ee6df3d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "339bba51d29c9df80d7ed456aaee4840f3eeab8b1db3cf0936c776b7847d3051", "generation_seconds_batch": 18.04512499505654, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6272985935211182, "target_z": -1.7773437917055177, "ce": 6.38797664642334, "genuineness": 4.143336772918701, "identity": 0.5226011872291565, "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/casual_not_formal/c771a65d208fde4288bf16dacbf4e9b17d72fe9109eedb6a30edde865ee6df3d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c7c2badf67cdb51795e0baab4a047b076325cbe780621b1ec14e6768796b7a64", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1fa48e4a8c017422aea9ee5e91a85fd24ac2bc19ed4d477023c809d8fc809e9b", "generation_seconds_batch": 22.59962709294632, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2844715118408203, "target_z": -0.5185546437791569, "ce": 6.070269584655762, "genuineness": 4.369274139404297, "identity": 0.7009772062301636, "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/casual_not_formal/c7c2badf67cdb51795e0baab4a047b076325cbe780621b1ec14e6768796b7a64.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "c7f153dcf72290a1c05bfbb6938c4ecf48aa17d7187595f012f3be90b8f118e7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d9c458ff12dbed0def780946a24a29814cc404de1a86d4d9e1703cd25271b4f", "generation_seconds_batch": 23.46962612506468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4795589447021484, "target_z": -0.6992187163069538, "ce": 5.959072113037109, "genuineness": 4.168440818786621, "identity": 0.44103169441223145, "hypothesis": "De deur staan nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/c7f153dcf72290a1c05bfbb6938c4ecf48aa17d7187595f012f3be90b8f118e7.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "50ee77708ee760b2f4676548c671253a076c9552c6863dd3b342aea1dcd0494d", "generation_seconds_batch": 17.258046666975133, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.593478202819824, "target_z": -0.33300776095076823, "ce": 6.308549880981445, "genuineness": 4.444586753845215, "identity": 0.7435764074325562, "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/casual_not_formal/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "c865abc12457437b44a652511b9589f0ff6ec0a88a6eec91439892e3376c7814", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c286ac603e81895d45cf16d328f0078ff4470f42f85f10b7e8d22fe0c61b05ef", "generation_seconds_batch": 24.368041634908877, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192267894744873, "target_z": -0.535156238469914, "ce": 6.229123115539551, "genuineness": 4.482243061065674, "identity": 0.842498242855072, "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/casual_not_formal/c865abc12457437b44a652511b9589f0ff6ec0a88a6eec91439892e3376c7814.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": "c874f4cfe0b39c4907e5ca916315a060cedd70be82114efec9b93a40aafc7d61", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cdb8307aa6c7b6b05200af938f44ab71da677bfbac0484c33f4e16e5e9f13e15", "generation_seconds_batch": 18.06424505810719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.161651611328125, "target_z": 0.10302729872715846, "ce": 6.133810997009277, "genuineness": 4.331617832183838, "identity": 0.6551401615142822, "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/casual_not_formal/c874f4cfe0b39c4907e5ca916315a060cedd70be82114efec9b93a40aafc7d61.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "c87a8aac2523484f35c779737ba2343d3dc22897bfdfe320b20b093673b3b074", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4eacc0ea7613ea900e4a6d0d68a627043513d80fc617e3674c8ec87c2c3e0c5e", "generation_seconds_batch": 26.539110217010602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.718790054321289, "target_z": -0.7656249656234385, "ce": 6.181467056274414, "genuineness": 4.344170093536377, "identity": 0.6877379417419434, "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/casual_not_formal/c87a8aac2523484f35c779737ba2343d3dc22897bfdfe320b20b093673b3b074.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "c198329920c034cd9447c1f22e2a9720008c62a7e7d35e67659f4d91a8a2d38c", "generation_seconds_batch": 17.1706567479996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.569270610809326, "target_z": -1.1425781692122663, "ce": 6.23706579208374, "genuineness": 4.118232727050781, "identity": 0.40611201524734497, "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/casual_not_formal/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.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": "c8ece4f4d2b6622be486e78f8499d715aad5ee144c4d2bb3221262d6c53fe53f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a89c5a4e2f676edb70303547d223475247611711b3c40328a2265472455174b7", "generation_seconds_batch": 18.675916413078085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.36919903755188, "target_z": -0.5107422517558575, "ce": 6.300607204437256, "genuineness": 4.15588903427124, "identity": 0.5762491226196289, "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/casual_not_formal/c8ece4f4d2b6622be486e78f8499d715aad5ee144c4d2bb3221262d6c53fe53f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "c91ef8cb6f19341a94b64674f84a043e1c761aaa1297f9a98b0eed490941cf5d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "35b82117b9820d312377366b80c4865adb5bd35158f81ba282a0387c0e5f7c03", "generation_seconds_batch": 17.501406223978847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0527164936065674, "target_z": -0.7578125001344612, "ce": 6.340320587158203, "genuineness": 4.494794845581055, "identity": 0.34282276034355164, "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/casual_not_formal/c91ef8cb6f19341a94b64674f84a043e1c761aaa1297f9a98b0eed490941cf5d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "c9411a6a1aa5fe6f855fbffc1771036c285e1f46622d854be8bd649ab021d23f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9bc0c0a191166ef1a5d62b00ca5504071d8b2e3cc7e6ab348e5dafbe5b5d43f", "generation_seconds_batch": 22.768299942021258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5685582160949707, "target_z": -0.1284180150514264, "ce": 6.197352409362793, "genuineness": 4.394378662109375, "identity": 0.6984350681304932, "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/casual_not_formal/c9411a6a1aa5fe6f855fbffc1771036c285e1f46622d854be8bd649ab021d23f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "c96b847b14acea86ee72946ff6b0fa71352b81c4c25afaf2f3fc93475e3d815a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87633b42d6b6dbb26758b237b1a68e765edb3145243d802379151f9cad0beb1b", "generation_seconds_batch": 23.456846726941876, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.615906000137329, "target_z": -0.18066407725836664, "ce": 6.046441555023193, "genuineness": 4.6203155517578125, "identity": 0.7759285569190979, "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/casual_not_formal/c96b847b14acea86ee72946ff6b0fa71352b81c4c25afaf2f3fc93475e3d815a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "c98df38fb914176922d6aeb0b092bf8933cea08100d0f54644d748fc51cd4402", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f0ea1d26233268a5fccb29cfc15b07593866ba68d6070f9e462437b5bf3630ed", "generation_seconds_batch": 18.648007082985714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0758566856384277, "target_z": -1.2968749921948208, "ce": 6.260893821716309, "genuineness": 4.306513786315918, "identity": 0.3619672954082489, "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/casual_not_formal/c98df38fb914176922d6aeb0b092bf8933cea08100d0f54644d748fc51cd4402.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "ca7a22a725b79621d180566d98c423f85b353e316c7c61f7eb4ad51299204240", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "093465effbb1618ff5d7c6f9dc61ccb581a931370b66f54432e54b6277fbd112", "generation_seconds_batch": 23.33904941997025, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.137444019317627, "target_z": -0.6582031135019304, "ce": 6.125868320465088, "genuineness": 4.432034492492676, "identity": 0.626234769821167, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/ca7a22a725b79621d180566d98c423f85b353e316c7c61f7eb4ad51299204240.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "caf1d6799d2b9e12bc888df7174cc1787f104dfa1cd350f64191adcb9bc72246", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f02bfeda402dd7a995802410d924ba532b3c06342209721a87a13c9d044bc1fe", "generation_seconds_batch": 15.289207939989865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1025562286376953, "target_z": -0.7734374812217921, "ce": 6.173524379730225, "genuineness": 4.532451152801514, "identity": 0.7123041749000549, "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/casual_not_formal/caf1d6799d2b9e12bc888df7174cc1787f104dfa1cd350f64191adcb9bc72246.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "cbb2a57e7f9c238b433e17e7d737fd0ae26c87d50835fb93525c759f51801f37", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "828569601a66bea5b276cca658f3654756a9257c97fd47706b8d4a7856772bf4", "generation_seconds_batch": 24.95866973802913, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.8320312773805523, "ce": 6.23706579208374, "genuineness": 4.425758361816406, "identity": 0.47766634821891785, "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/casual_not_formal/cbb2a57e7f9c238b433e17e7d737fd0ae26c87d50835fb93525c759f51801f37.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "cbe8c974f42f9e2ae292d2aedbcdda7ef2dde19b562aa941f90caf5040f7be3b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1626b1415057f1558a39323693949bf7a1ebce09f350a852725fffb43e341fe", "generation_seconds_batch": 17.66224496497307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.524414539337158, "target_z": -1.2226563526166443, "ce": 6.173524379730225, "genuineness": 4.331617832183838, "identity": 0.6975917816162109, "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/casual_not_formal/cbe8c974f42f9e2ae292d2aedbcdda7ef2dde19b562aa941f90caf5040f7be3b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "cbea31bf9fbd15837dcfed442b3823d1bfdcd2a0c09f32113838b3b8cd46be40", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1c8f903f6fc848854b28a96adaf71d00fb8c810f4fd271774ea80bff35597fa", "generation_seconds_batch": 15.564135294989683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.192267894744873, "target_z": -0.6406250466581888, "ce": 5.871703147888184, "genuineness": 4.356721878051758, "identity": 0.7531064748764038, "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/casual_not_formal/cbea31bf9fbd15837dcfed442b3823d1bfdcd2a0c09f32113838b3b8cd46be40.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": "cca9d9559c42c1ec44af01da1cc08dcd362268091263fd4bee4997e170fe6c54", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "5461578afd32ce0564b98a32aa5601dc857413994ad3654f68745d1801a7dca2", "generation_seconds_batch": 18.035754844895564, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411562919616699, "target_z": -0.5859375412797238, "ce": 5.895530700683594, "genuineness": 4.570107460021973, "identity": 0.7764684557914734, "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/casual_not_formal/cca9d9559c42c1ec44af01da1cc08dcd362268091263fd4bee4997e170fe6c54.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "ccdeed6d352887736b6ac7bd643ce3033c7f408e85067979b89415468f7b19c1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b99316316e984f017018d01c74f621e34844ef710cc6a8e271cfc1387067306", "generation_seconds_batch": 17.63638815190643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9430689811706543, "target_z": -0.6914062602126809, "ce": 6.197352409362793, "genuineness": 4.425758361816406, "identity": 0.6615060567855835, "hypothesis": "Die Tür ist noch offen. Ich hab 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/casual_not_formal/ccdeed6d352887736b6ac7bd643ce3033c7f408e85067979b89415468f7b19c1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "cce467ae89d044d385ab6b9354a7fc19b69645f10008bec2970b24d902063533", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9c8d9429e5e70ab6d15e5b17ced76976aa840276a6d7ea45c4ecedcf565495e7", "generation_seconds_batch": 17.12671903090086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.024592399597168, "target_z": 0.03222660093001811, "ce": 6.070269584655762, "genuineness": 4.482243061065674, "identity": 0.20667245984077454, "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/casual_not_formal/cce467ae89d044d385ab6b9354a7fc19b69645f10008bec2970b24d902063533.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "cd4da6a9133f55cbc9fb70ec2ac5c8aca25a76333def6a8a2c9f5cfabdee219b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4c8e09261f5e5cefac6c0d9a4470994fb8905aab7c363d7f7a2b7f633965525b", "generation_seconds_batch": 21.84458421205636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9961129426956177, "target_z": -1.3828125150757113, "ce": 6.149696350097656, "genuineness": 4.243753433227539, "identity": 0.6481653451919556, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hin.", "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/casual_not_formal/cd4da6a9133f55cbc9fb70ec2ac5c8aca25a76333def6a8a2c9f5cfabdee219b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "ce562a98c20ac0fd4ef872e95db83b409bdeda74a4349dd3623e21c64b55d9da", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6766fa35ca0141fe28a877e7bf6add0320db5f8b1d8da3c4e07fc9c9eb203ef3", "generation_seconds_batch": 16.88869151403196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9131650924682617, "target_z": -0.8749999536107336, "ce": 6.0305562019348145, "genuineness": 4.482243061065674, "identity": 0.6161161661148071, "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/casual_not_formal/ce562a98c20ac0fd4ef872e95db83b409bdeda74a4349dd3623e21c64b55d9da.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "ce6de8908a5c98f2c12a010e24bbd764dfc1d3e5685e6ba52734c7b16065b644", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2a5bc148986dfbdac7506d65735f3f1c7cb9ecbf886be707f3db4e721079c024", "generation_seconds_batch": 21.607974275015295, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.4218749216648878, "ce": 5.879645347595215, "genuineness": 4.419482707977295, "identity": 0.5694320201873779, "hypothesis": "Die Tür ist noch offen. Ich habe dein 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/casual_not_formal/ce6de8908a5c98f2c12a010e24bbd764dfc1d3e5685e6ba52734c7b16065b644.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "cf10b262e6b8149fd8068b4e9c99a8c4aaca0ab7998ece82249d2057a990bbb8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "89c1fca2c04ff56a2fbaf9915e00efc18af18f0078f7445b19a5c3bedf1e37eb", "generation_seconds_batch": 25.88203698990401, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.6835937354157142, "ce": 6.181467056274414, "genuineness": 4.469690799713135, "identity": 0.6832373142242432, "hypothesis": "La porte est encore ouverte, j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/cf10b262e6b8149fd8068b4e9c99a8c4aaca0ab7998ece82249d2057a990bbb8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "cf5a04de13770a13b49f2161304db494ff69ae6a866e0878681c38821161d1d5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "3b4028c83fb2dcebfc63a3bf8b7fe5c67757a939d56efcc5e6259a279b976e71", "generation_seconds_batch": 19.9915918110637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0078606605529785, "target_z": -0.6464843724340237, "ce": 5.80021858215332, "genuineness": 4.118232727050781, "identity": 0.6982296705245972, "hypothesis": "La puerta sigue abierta. He dejado tu abrigo sobre la silla. Dime cuándo está abierto.", "wer": 0.13333333333333333, "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/casual_not_formal/cf5a04de13770a13b49f2161304db494ff69ae6a866e0878681c38821161d1d5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "cf640b6db329dac330c97d249ce293f8a635a106c19d99073a7a8dbb8a62ccd8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fb4d8ccf454b7e7dc06bfe83303243f52de15f64834811557add322f49a6f92", "generation_seconds_batch": 20.963755113072693, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4247350692749023, "target_z": -1.1992188553442928, "ce": 6.427689552307129, "genuineness": 4.18099308013916, "identity": 0.6375647783279419, "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/casual_not_formal/cf640b6db329dac330c97d249ce293f8a635a106c19d99073a7a8dbb8a62ccd8.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "3448117ba78610b564c9178830e673c4d991e8c58a935ba39668c94667abe237", "generation_seconds_batch": 15.827006644918583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3150875568389893, "target_z": -1.4335938146232254, "ce": 6.395919322967529, "genuineness": 4.482243061065674, "identity": 0.5651070475578308, "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/casual_not_formal/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "cf77fdd7a49947467fbabfe8bfc4e19af5fe1a437a1e7466fa408f9980ec4996", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3aff683ac78a1f88aac25b6ddcaaac214b32603693f88bb7133544e607027cf4", "generation_seconds_batch": 23.91885087499395, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5286865234375, "target_z": -0.09570308322764946, "ce": 5.95112943649292, "genuineness": 4.432034492492676, "identity": 0.7975705862045288, "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/casual_not_formal/cf77fdd7a49947467fbabfe8bfc4e19af5fe1a437a1e7466fa408f9980ec4996.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": "cf9c6945f6da2147fe7296ea882b4e4d6d3f70ca141161fa1ae6e9c8bb6286c2", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b9a306c3109b6323d4dd232fd682b28965c4a77b1c6f8f6d5cb57a60bcf81c5", "generation_seconds_batch": 24.031974275945686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5648202896118164, "target_z": -0.3071289169228926, "ce": 6.157639026641846, "genuineness": 4.406930446624756, "identity": 0.6811255216598511, "hypothesis": "La carretera gira en la izquierda tras el puente. Sigue hasta la pequeña puerta de madera. Te esperaré al otro lado.", "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/casual_not_formal/cf9c6945f6da2147fe7296ea882b4e4d6d3f70ca141161fa1ae6e9c8bb6286c2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "cfab0844c341a1c08211d22f598456fb7b9e9ab382971708a5c5fb5500cee768", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9478149edc89525ff708a2f74fa4d3abf2adcc5656e6c41b29d98e5356c227e7", "generation_seconds_batch": 25.70433899492491, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.863325357437134, "target_z": -0.7851562012800645, "ce": 6.300607204437256, "genuineness": 4.15588903427124, "identity": 0.14322249591350555, "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/casual_not_formal/cfab0844c341a1c08211d22f598456fb7b9e9ab382971708a5c5fb5500cee768.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "d00e0ac7e7b519ba01b2238dc54c6947cb709202b50383278e432a823f846bce", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "052cb49e977ed38bf31e82055927a331e421e00a4d1326ed25e65e67eb283f2c", "generation_seconds_batch": 17.611180773936212, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.790968418121338, "target_z": -0.12670895842030044, "ce": 6.340320587158203, "genuineness": 4.469690799713135, "identity": 0.3134300112724304, "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/casual_not_formal/d00e0ac7e7b519ba01b2238dc54c6947cb709202b50383278e432a823f846bce.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": "d07487836a1e4d206505b75034c875f2f3a7093c5ccb6e315ba3cf3071c92ff5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5774f55cc5dfc4bf4338bfeb0121b693f302f4c930bc7959fc03c385da927759", "generation_seconds_batch": 24.307468662038445, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.547910213470459, "target_z": 0.6074219209562689, "ce": 6.149696350097656, "genuineness": 4.557555198669434, "identity": 0.6921293139457703, "hypothesis": "De weg gaat linksaf naar de brug. Loopt verder tot aan het kleine houtenhek. Ik wacht aan de andere kant.", "wer": 0.19047619047619047, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/d07487836a1e4d206505b75034c875f2f3a7093c5ccb6e315ba3cf3071c92ff5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "d07f3f0529dfc567d8f51b02840219b1b41b24b088ad2da49abcf7b2012f68f9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "67493d314e09bbf8200531524da22edb54b66f96b5fd8b4de9472a370bd91e84", "generation_seconds_batch": 20.710037415032275, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2154078483581543, "target_z": -1.332031288864219, "ce": 6.117925643920898, "genuineness": 4.344170093536377, "identity": 0.6500282883644104, "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/casual_not_formal/d07f3f0529dfc567d8f51b02840219b1b41b24b088ad2da49abcf7b2012f68f9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "d08b1d5a4ada569e53aaa3f44edb7dcdb9633312729386c9a739cbf46dd0a0a2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0d2e52c8956c7dfa329e747a608720adc546a72b06afd5a991c30a4e2d44abff", "generation_seconds_batch": 17.501406223978847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.889401912689209, "target_z": 0.039794900589242226, "ce": 6.09409761428833, "genuineness": 4.400654315948486, "identity": 0.8127113580703735, "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/casual_not_formal/d08b1d5a4ada569e53aaa3f44edb7dcdb9633312729386c9a739cbf46dd0a0a2.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "1bdfa9fe508312bdba3de5cef1503d68f423db17484b33d21f8f81a1b8be1f4f", "generation_seconds_batch": 17.573292694054544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.888245105743408, "target_z": -0.9511719212660101, "ce": 6.41180419921875, "genuineness": 4.231201171875, "identity": 0.16505387425422668, "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/casual_not_formal/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.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": "d2130ed29ceb86b4ebdf51ae88a56a1a2f8e25deb69ae671f78a70a0d74e0576", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "87bd4347ed5b7aeca4491d4020c93bc46a204367d1b22420441a270cf4ee157e", "generation_seconds_batch": 16.859064204036258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4165468215942383, "target_z": -0.2871094205280725, "ce": 6.054384231567383, "genuineness": 4.413206577301025, "identity": 0.6167964935302734, "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/casual_not_formal/d2130ed29ceb86b4ebdf51ae88a56a1a2f8e25deb69ae671f78a70a0d74e0576.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "4d182e25c16150397eab41a48c90aee0d9b1c03d031f2f237d3ee6c9ac46cabf", "generation_seconds_batch": 15.598526270012371, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.5039062594619352, "ce": 6.23706579208374, "genuineness": 4.080576419830322, "identity": 0.748336911201477, "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/casual_not_formal/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "d27cac8e502032beac5ede6912ce9e66c06871b163f4b7ef7bddf089289169e8", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07dbc659583827b5754399595e5e36db279395beaeeb4ecf53a2022e7ad1764e", "generation_seconds_batch": 17.510748381959274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.098996877670288, "target_z": -2.085937543367079, "ce": 6.205295085906982, "genuineness": 3.9425034523010254, "identity": 0.6949774026870728, "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/casual_not_formal/d27cac8e502032beac5ede6912ce9e66c06871b163f4b7ef7bddf089289169e8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "d27e4d654e736fbdf13db62071e7adf0a6b53387bcbe607537070f58d6949fd5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9604d7e1b5145ba8bc1205b5878993c60ac8005e102b73ba9a7ffe9a4f142819", "generation_seconds_batch": 21.010590536054224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.148437537658865, "ce": 6.141753673553467, "genuineness": 4.293961524963379, "identity": 0.5247902870178223, "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/casual_not_formal/d27e4d654e736fbdf13db62071e7adf0a6b53387bcbe607537070f58d6949fd5.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": "d290b4c7643bc0a912583f9e2208fbe996d30c16984d17af924cccd6d3822678", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c153aa208ffe5008e6486316566cd86f805c3e88faf107fd5d717e4ba9a8f16", "generation_seconds_batch": 18.06424505810719, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5212104320526123, "target_z": -0.3164062534928018, "ce": 6.141753673553467, "genuineness": 4.331617832183838, "identity": 0.18678319454193115, "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/casual_not_formal/d290b4c7643bc0a912583f9e2208fbe996d30c16984d17af924cccd6d3822678.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "d3d43853902e2d64a45a40e7a021057105c723ffc80c5d81c6e6a9adfd86da17", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b943fc16317fa509a2e3361436bc664647d81b9768bfa5aef6581385b6a66249", "generation_seconds_batch": 24.1828153639799, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5536062717437744, "target_z": -0.39746100156617026, "ce": 6.213237762451172, "genuineness": 4.457139015197754, "identity": 0.48109519481658936, "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/casual_not_formal/d3d43853902e2d64a45a40e7a021057105c723ffc80c5d81c6e6a9adfd86da17.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "9e5a78a238b3e138c652ca482f3a760302143f860950341e3690370a417aedf5", "generation_seconds_batch": 17.182745361002162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.51444673538208, "target_z": -1.2968749629654228, "ce": 6.133810997009277, "genuineness": 4.582659721374512, "identity": 0.7488539814949036, "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/casual_not_formal/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "d46dfb462b011684c719af6fed2b669d6b1a566dbe7b7ec6004344cbef1b111f", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "698a3c9c833c5c09e4fcbc511e2cbf378458e0a612117ee7535132fe18bfbd36", "generation_seconds_batch": 16.485142056946643, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3442792892456055, "target_z": -0.49121089891911274, "ce": 6.221180438995361, "genuineness": 4.607763767242432, "identity": 0.40230679512023926, "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/casual_not_formal/d46dfb462b011684c719af6fed2b669d6b1a566dbe7b7ec6004344cbef1b111f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "d4ba2bc623f9b2530a4254971a6b63b52cd4faedebad9aec25a9786f114f7d9c", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b495d9b0a7e94581ddabb287cef0f94f48f0ca6d2dfaabb67495d626aa622df", "generation_seconds_batch": 15.738997455919161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.673933982849121, "target_z": -0.9609375383439791, "ce": 6.125868320465088, "genuineness": 4.469690799713135, "identity": 0.8681097030639648, "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/casual_not_formal/d4ba2bc623f9b2530a4254971a6b63b52cd4faedebad9aec25a9786f114f7d9c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "d4bf30f3f075a61b3c0de68969176e46704f6415709327a2c37a062c9dc3e09e", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7491022221819d74dac5121677a7e06a0a0253eaefef00fd99d496c9216c6ca2", "generation_seconds_batch": 17.855723098036833, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -0.7080078498471207, "ce": 6.23706579208374, "genuineness": 4.7834930419921875, "identity": 0.5929749011993408, "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/casual_not_formal/d4bf30f3f075a61b3c0de68969176e46704f6415709327a2c37a062c9dc3e09e.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": "d56324149e761138eeae5b8e5ca4f122ebd927c019aa341f33764a1dc4016953", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1d6db02a8ba0fe215fdaca93a519b1fd8aa5a92efa37c6df7386ee8375433a41", "generation_seconds_batch": 15.840128924930468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6632537841796875, "target_z": -0.221679706390261, "ce": 6.427689552307129, "genuineness": 4.325342178344727, "identity": 0.3296867609024048, "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/casual_not_formal/d56324149e761138eeae5b8e5ca4f122ebd927c019aa341f33764a1dc4016953.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "d565ebed5e389f3686410e97d6b7c7edaf48da179171616cb881ee09f9e6bcd9", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1038278dd4200a10ba4375d478c4f818208f864a7f085be40018fdd11ea00efa", "generation_seconds_batch": 17.60875292599667, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.092588424682617, "target_z": -0.7539062762952523, "ce": 5.982900142669678, "genuineness": 4.444586753845215, "identity": 0.6939366459846497, "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/casual_not_formal/d565ebed5e389f3686410e97d6b7c7edaf48da179171616cb881ee09f9e6bcd9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "d58477d8256bfdeecd12fcd8a1744425be0abe39b254bff32f38242ae1a72e5f", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18a80f1c6119cde26d9d61e2f32b0dc904606e8a2622a8e94cd6f2ddd4065b08", "generation_seconds_batch": 14.930282880901359, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8433892726898193, "target_z": -0.8554687233701936, "ce": 6.117925643920898, "genuineness": 4.469690799713135, "identity": 0.6473209857940674, "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/casual_not_formal/d58477d8256bfdeecd12fcd8a1744425be0abe39b254bff32f38242ae1a72e5f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "d59c8d4f60e0814075df3ef074fad8f283bace762748234f5233a77069507a45", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "276c2dc70487b368b3b183dd4a064add0b318b62d64e8d3669e5fd21a10f13e9", "generation_seconds_batch": 22.496173362014815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4379072189331055, "target_z": -2.03125001424013, "ce": 6.372091293334961, "genuineness": 4.105680465698242, "identity": 0.6320067644119263, "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/casual_not_formal/d59c8d4f60e0814075df3ef074fad8f283bace762748234f5233a77069507a45.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "d5dce9d5dad2df48b0c59deb846c2d876c3996cf83f6572bcd8513d411aebe85", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "db81962e9ad68d468de43d0ee68c55cf8d3595acaa5db1edb6bdc7ba539da9cb", "generation_seconds_batch": 15.448804730083793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.883261203765869, "target_z": -0.7929687763160636, "ce": 5.967014789581299, "genuineness": 4.519899368286133, "identity": 0.5962002277374268, "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/casual_not_formal/d5dce9d5dad2df48b0c59deb846c2d876c3996cf83f6572bcd8513d411aebe85.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "d650a6456d39cacf646d74cba42584df52fd2b88b80c7d7f69d94a15835a1605", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6d68affa5002df17336d7075a0d6997ab96cf21f91f1fac27f79e3ad567f853e", "generation_seconds_batch": 23.00830056006089, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.305119514465332, "target_z": -1.3242187656327684, "ce": 6.197352409362793, "genuineness": 4.218649387359619, "identity": 0.6768285036087036, "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/casual_not_formal/d650a6456d39cacf646d74cba42584df52fd2b88b80c7d7f69d94a15835a1605.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "d653d88e7ff03da61884d07ad6b24d30c8ef1ee9df3e57f52378243d08bccef5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "52a5867c23684c29c3e7c2c6f674e71c92dff46f48e1e3af752b4562da9aed6e", "generation_seconds_batch": 17.841417910065502, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5025205612182617, "target_z": -0.3139648125228847, "ce": 5.8875885009765625, "genuineness": 4.532451152801514, "identity": 0.7501106262207031, "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/casual_not_formal/d653d88e7ff03da61884d07ad6b24d30c8ef1ee9df3e57f52378243d08bccef5.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": "d659874b5ff626d58776982361f57b3e4e3ee5bf0028debd9892ba10d85b236f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 128, "seconds_generated": 10.24, "mp3_sha256": "bdd4dfac58802bab0b23f97d9f6f3ca320dc11829e12fb83eb457a8c2f6a09a1", "generation_seconds_batch": 16.859064204036258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.313663005828857, "target_z": 0.35253899292050184, "ce": 6.117925643920898, "genuineness": 4.582659721374512, "identity": 0.6409472823143005, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu et puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "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/casual_not_formal/d659874b5ff626d58776982361f57b3e4e3ee5bf0028debd9892ba10d85b236f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "d66f11fad05b36da99b6aa76930b93f1e57ff6f564e2157d7ca641d143b5dd59", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87349e23927e263f1f8008fe4547b0b5fa32e30baa9b376e29796bbfa2f01619", "generation_seconds_batch": 17.55643746908754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5797722339630127, "target_z": -0.02392579343878745, "ce": 6.117925643920898, "genuineness": 4.356721878051758, "identity": 0.49247732758522034, "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/casual_not_formal/d66f11fad05b36da99b6aa76930b93f1e57ff6f564e2157d7ca641d143b5dd59.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "d6735fcd69a0f20c8adcb11a8ff3648688608ac8cd4d680838300b18b4d718d0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5707c14952aecad7728b808030aae7e44653cd2427c00798590311da8b3d388d", "generation_seconds_batch": 18.291450777091086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.035984754562378, "target_z": -1.5625000907231898, "ce": 6.268836498260498, "genuineness": 4.281409740447998, "identity": 0.542304515838623, "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/casual_not_formal/d6735fcd69a0f20c8adcb11a8ff3648688608ac8cd4d680838300b18b4d718d0.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": "d74a4bb04e511c8be400164623037c9450b6b7f6391cd386ef3aae162f739ab5", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c3946aef928138aa827829727e8c64e59cfcb26316d335b7e6e2180f3e001f66", "generation_seconds_batch": 17.47923596005421, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1623640060424805, "target_z": -0.7070312936263992, "ce": 6.23706579208374, "genuineness": 4.362998008728027, "identity": 0.6868300437927246, "hypothesis": "Te escribí el número en el sobre. Léelo bien y comprueba las dos últimas cifras. Llámame cuando tengas un momento.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/d74a4bb04e511c8be400164623037c9450b6b7f6391cd386ef3aae162f739ab5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "d75489be4f56da5789ebea65db955c2ad3960fd2204177f7abbdd9ae8436dbd0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a0edec2bf5996ce664c14f961ec42cfb7527b7f43dc324af793622ef5ae7d379", "generation_seconds_batch": 23.46962612506468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0260167121887207, "target_z": -1.246093776031135, "ce": 6.213237762451172, "genuineness": 4.406930446624756, "identity": 0.5859889984130859, "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/casual_not_formal/d75489be4f56da5789ebea65db955c2ad3960fd2204177f7abbdd9ae8436dbd0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "d7eee6b7e6897a175a3104e0308e0fef1701cbcb2ad0dcf5cfbeaa842bdd8923", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab1a4a0dc920622844767ccffe71a335b3bd00a05692d448fe3f975084be3f97", "generation_seconds_batch": 17.054144437075593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5991744995117188, "target_z": -1.1269530623328705, "ce": 6.260893821716309, "genuineness": 4.457139015197754, "identity": 0.596908688545227, "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/casual_not_formal/d7eee6b7e6897a175a3104e0308e0fef1701cbcb2ad0dcf5cfbeaa842bdd8923.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": "d7fc0a1f95c8e7831b6e78025912c5d9c4a16d9b202f863e890d84d3dd7e9449", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9a816107d707283e40da1eb1234c9cbdebc36e142b9bbc4e0b478d4568fb8e6d", "generation_seconds_batch": 16.22309313202277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2154078483581543, "target_z": -1.4062500601075572, "ce": 6.260893821716309, "genuineness": 4.168440818786621, "identity": 0.5411257743835449, "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/casual_not_formal/d7fc0a1f95c8e7831b6e78025912c5d9c4a16d9b202f863e890d84d3dd7e9449.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "d82f084bb79c2eb7bd3acb8561185901cf50f2df81f02029242d225aab7cc08d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "969420117a853833f31de0f197680298a06b2ccb00679060abf75e1a187cda59", "generation_seconds_batch": 16.845799711998552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.717010259628296, "target_z": -1.7343749399660775, "ce": 6.316492557525635, "genuineness": 4.017816066741943, "identity": 0.5577950477600098, "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/casual_not_formal/d82f084bb79c2eb7bd3acb8561185901cf50f2df81f02029242d225aab7cc08d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "d83a9a89a16408f20d3bb2fc3c1e5cc88f3d724abecf4a18d6290411dd9c9a4c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e966e0859b3e7e9ddb4671ba10f963ae441b6f4d850fd69d3ce39d8f034fe3da", "generation_seconds_batch": 25.666605120990425, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7049946784973145, "target_z": -0.08032228594591574, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.8007006049156189, "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/casual_not_formal/d83a9a89a16408f20d3bb2fc3c1e5cc88f3d724abecf4a18d6290411dd9c9a4c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "d885af551850559d629ef8c0c984ea5e9fb9bee601e65168f4949e779cb22cb2", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf2d56fd2ae1767fa99390c2a7efd0627f1bbf3bff847ec76a0a0dc959e824cc", "generation_seconds_batch": 16.973842874052934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1556003093719482, "target_z": -1.183593770647865, "ce": 6.260893821716309, "genuineness": 4.13078498840332, "identity": 0.23403260111808777, "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/casual_not_formal/d885af551850559d629ef8c0c984ea5e9fb9bee601e65168f4949e779cb22cb2.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": "d8c6173b13c0d51fa27caa0e2a80ada9933a5d5605f1f7b17a8a4fe895800a9a", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "0e0ca2e8c79b6de011363de154ca25d4a861fe813e213e52452d2855dd7d8684", "generation_seconds_batch": 17.34960559103638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4090709686279297, "target_z": -0.5263671901724248, "ce": 6.070269584655762, "genuineness": 4.507347106933594, "identity": 0.74260014295578, "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/casual_not_formal/d8c6173b13c0d51fa27caa0e2a80ada9933a5d5605f1f7b17a8a4fe895800a9a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "d8d16bc9e861d7034af9690312a7b065c3fc508ac7966640044310a2bfa643a8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9b6ecc4f3eec1c6af3156eedac2cb7f088c468ce739d36d941feab694b56464a", "generation_seconds_batch": 24.86681206803769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.977956771850586, "target_z": -0.6679687688782543, "ce": 6.141753673553467, "genuineness": 4.300237655639648, "identity": 0.6799584627151489, "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/casual_not_formal/d8d16bc9e861d7034af9690312a7b065c3fc508ac7966640044310a2bfa643a8.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": "d908a71aa1aeebc4d79e3a327819f9f6d84875692b3899d3ef79e4a7ad8512eb", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "defb71e31b62ee6fc9bfeb2ae7771b13ca100cc73bcd2e3a337c2457855f66a1", "generation_seconds_batch": 16.110939012025483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.577280282974243, "target_z": -0.29052732980040447, "ce": 6.340320587158203, "genuineness": 4.105680465698242, "identity": 0.38907110691070557, "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/casual_not_formal/d908a71aa1aeebc4d79e3a327819f9f6d84875692b3899d3ef79e4a7ad8512eb.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "d917b0b89a1d6c503af3cf363047786f0a088ab6d1c021de4d5c30d31ede4b88", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "39affb4ac5ec6f1f68556103ec6be790a4dc55f94c39f65edba9794d88141bfa", "generation_seconds_batch": 17.38358975807205, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.982940673828125, "target_z": -0.7148437569631935, "ce": 6.062326908111572, "genuineness": 4.406930446624756, "identity": 0.7372583746910095, "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/casual_not_formal/d917b0b89a1d6c503af3cf363047786f0a088ab6d1c021de4d5c30d31ede4b88.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": "d929436c357a62be62147a4a838384a54b70202fe9461f06cfd62d0a73570fee", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7fa219e565cd36648d377e68f429dbbd5c3678e7b7e2bd01923b77f4383a7c3", "generation_seconds_batch": 21.94623415195383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.446450710296631, "target_z": -0.4628906495456535, "ce": 6.300607204437256, "genuineness": 4.300237655639648, "identity": 0.4265422821044922, "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/casual_not_formal/d929436c357a62be62147a4a838384a54b70202fe9461f06cfd62d0a73570fee.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "d933ad9d1b619d1f817973bc5205eea7f618eca870868eb2045f8b05842766a1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f8e477e4e30572838530dc5d383faaf9a5a149da26f5d0b7e6f12258b569d2b0", "generation_seconds_batch": 20.420893021975644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.147412061691284, "target_z": -0.5351562085225858, "ce": 6.284721851348877, "genuineness": 4.137060642242432, "identity": 0.1923500895500183, "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/casual_not_formal/d933ad9d1b619d1f817973bc5205eea7f618eca870868eb2045f8b05842766a1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "d93f1d70f8adc9f446019652538243859f5a1f1b597d522c854eac20467b6012", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6bef9d28ebb8002f8ea3de717b03e37e674af56dc068c115612caf37c25b1df3", "generation_seconds_batch": 23.79364049504511, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7337417602539062, "target_z": -0.757812505615375, "ce": 6.141753673553467, "genuineness": 4.519899368286133, "identity": 0.6091823577880859, "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/casual_not_formal/d93f1d70f8adc9f446019652538243859f5a1f1b597d522c854eac20467b6012.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "d9b07fede6fce0443f88fa05df9bf0a61493f28674c78afb2368e251038e4927", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a3bf169254d3f9ee034d62cfaa0ebc39e71ccdc0a172a62c676c983b1fda6c0", "generation_seconds_batch": 25.887744396924973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6775827407836914, "target_z": -0.1535644503713387, "ce": 6.022613525390625, "genuineness": 4.268857479095459, "identity": 0.5363789200782776, "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/casual_not_formal/d9b07fede6fce0443f88fa05df9bf0a61493f28674c78afb2368e251038e4927.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "da2781eccf686f504a71de41cc66c351c8142a9ba043222f4bc82acde0e8f980", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bae49c3b438f3a9b0ae3f7de15c69cdd7585c33f2236166cca1d08b04e6c97c", "generation_seconds_batch": 17.240946538979188, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6871063709259033, "target_z": -1.832031164767326, "ce": 6.467403411865234, "genuineness": 4.231201171875, "identity": 0.5991199016571045, "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/casual_not_formal/da2781eccf686f504a71de41cc66c351c8142a9ba043222f4bc82acde0e8f980.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "da664ed836c57671a6aeb97a990da448dd6ffba7e924c044b251b946a1bbfa80", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "035c9aac073071d2b69a360375fc68d1d4ee69652aeb016167aa769df2133009", "generation_seconds_batch": 22.252327944966964, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.6406248847199516, "ce": 5.911416053771973, "genuineness": 4.419482707977295, "identity": 0.5066661238670349, "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/casual_not_formal/da664ed836c57671a6aeb97a990da448dd6ffba7e924c044b251b946a1bbfa80.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "da98c8b2afcfe08adb617d2ff4207b52e66f1a7c711ffa34e0ca46a06036b69e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "795a0246c17e1663d082543bed37883bbb1067995b36cb115f46f5130de4b31c", "generation_seconds_batch": 15.564135294989683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3392951488494873, "target_z": -0.528320415748933, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.6651877760887146, "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/casual_not_formal/da98c8b2afcfe08adb617d2ff4207b52e66f1a7c711ffa34e0ca46a06036b69e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "dabc1a7a348b2c653e23f118cdd4ead1678d1975e95f97d29baeadcc87cee204", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "743643bfc00c84daef00d4e3001932c679875749eba9a96a0dc24b8871b8fb5e", "generation_seconds_batch": 15.352219147025608, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3250555992126465, "target_z": -1.3124999667575605, "ce": 6.4912309646606445, "genuineness": 4.394378662109375, "identity": 0.585963249206543, "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/casual_not_formal/dabc1a7a348b2c653e23f118cdd4ead1678d1975e95f97d29baeadcc87cee204.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "aa38e5d1ba71b102732d8aeb3c217707fc20531239b7eaa09566287e165e6363", "generation_seconds_batch": 15.827717834035866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4247350692749023, "target_z": -1.0312500323780438, "ce": 6.419747352600098, "genuineness": 4.250029563903809, "identity": 0.33057475090026855, "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/casual_not_formal/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.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": "db4146d7c69de3910a7c8c6326662560f183121d7eb84ffa29bec50d7d6dea6b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "23797e465018353ffe256e7ef4394818f3e318d2020e3b7dbb78125c6e4b5f78", "generation_seconds_batch": 24.236826284090057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9436025619506836, "target_z": 0.01367179767515092, "ce": 6.181467056274414, "genuineness": 4.6203155517578125, "identity": 0.7065774202346802, "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/casual_not_formal/db4146d7c69de3910a7c8c6326662560f183121d7eb84ffa29bec50d7d6dea6b.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": "db547c8a72618705aebc282ad2188cf463c1aac64372708264aec6290568d216", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I wrote the number on the envelope.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 128, "seconds_generated": 10.24, "mp3_sha256": "5b762701913c7ff2c04a655965d009bdf680872a61c90e392fa39c91c0f378f6", "generation_seconds_batch": 18.7826644720044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8950088024139404, "target_z": -0.01269535842305334, "ce": 6.292664527893066, "genuineness": 4.218649387359619, "identity": 0.3989546000957489, "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/casual_not_formal/db547c8a72618705aebc282ad2188cf463c1aac64372708264aec6290568d216.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "dbbc4b5a52ed58815ee641439f7bb6d982b261ff966b6a969b687f88ae659552", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9378218d5b5aa0123261dc87ea7d89844b3cf2ae25b3eeb514633736cf022727", "generation_seconds_batch": 23.014348095981404, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4090709686279297, "target_z": -0.3916015203766895, "ce": 5.919358730316162, "genuineness": 4.570107460021973, "identity": 0.5536576509475708, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/dbbc4b5a52ed58815ee641439f7bb6d982b261ff966b6a969b687f88ae659552.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "dbc9d7864e0b21b0625283540b00d8ca7b50199759897a725de8222606d07011", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "752ef7646281a5667b0b083d293be0f88b8da59de87dd234ac5622cbdea633ce", "generation_seconds_batch": 18.71418038289994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.833421468734741, "target_z": -0.7460936991718983, "ce": 5.895530700683594, "genuineness": 4.344170093536377, "identity": 0.7034645080566406, "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/casual_not_formal/dbc9d7864e0b21b0625283540b00d8ca7b50199759897a725de8222606d07011.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "dbf2e68f5bcb410d0759a84945289df62edb50646ebfe03e8f83f55be135ed9a", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6c2859c59904494085fe29b2bbb268dea102b091777bd2be9ac3dc819a904ce", "generation_seconds_batch": 21.182026260066777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.13673210144043, "target_z": 0.2275391706815964, "ce": 6.038498878479004, "genuineness": 4.381826400756836, "identity": 0.538974404335022, "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/casual_not_formal/dbf2e68f5bcb410d0759a84945289df62edb50646ebfe03e8f83f55be135ed9a.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": "dc06349344a2cf9a09b2e1895f6cb19320b30e5d7ce0d001b98d4b9a900c24f7", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8897d18c7fdfc47d2d8a53b50b3f0fed8609c8f1a580d3ebdeede8ce556a9070", "generation_seconds_batch": 16.02424073696602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.139224052429199, "target_z": 0.0737305626848648, "ce": 6.062326908111572, "genuineness": 4.545003414154053, "identity": 0.814516544342041, "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/casual_not_formal/dc06349344a2cf9a09b2e1895f6cb19320b30e5d7ce0d001b98d4b9a900c24f7.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": "dc2b4e0911cb5c5030841886d66ec2c4f0df96b0cdc0091666561fff49f50674", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "40377df263360bb7c48922728fdf29dfa5ca0991bc3259546873007c0a8f32ab", "generation_seconds_batch": 16.3283529530745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7486045360565186, "target_z": -0.11987303235114152, "ce": 6.41180419921875, "genuineness": 4.419482707977295, "identity": 0.26730042695999146, "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/casual_not_formal/dc2b4e0911cb5c5030841886d66ec2c4f0df96b0cdc0091666561fff49f50674.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "dc91185beae56c9a586918ebc497940dfdfadf70db9e786a71b7b98bcfd001b5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1c47f850c2ba9422c2e63d436649eb0a5a5e0982e4df5e10cd048186a3de82ff", "generation_seconds_batch": 24.37862454901915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.478846549987793, "target_z": -0.16699225183749522, "ce": 6.046441555023193, "genuineness": 4.319066047668457, "identity": 0.7271018028259277, "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/casual_not_formal/dc91185beae56c9a586918ebc497940dfdfadf70db9e786a71b7b98bcfd001b5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "dca8bcd243425619cb326f752cde9740b57e55105578fd095290bfbd36fdfb11", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b068c6e9bd4228fb5136d35e6162daf37672f1703f08cc15652035fb379db677", "generation_seconds_batch": 17.366269557969645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1755361557006836, "target_z": -1.2968750947602405, "ce": 5.816103935241699, "genuineness": 3.9801597595214844, "identity": 0.5768212080001831, "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/casual_not_formal/dca8bcd243425619cb326f752cde9740b57e55105578fd095290bfbd36fdfb11.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "dce0c9057eff22fc0583d20764d07e6713cfe1874c43a55d08c4f799fd179472", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df4cd550aa6989c8683ea145a88e3d0e31e289ce12dff2111566425ed37311c7", "generation_seconds_batch": 24.4971581940772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.104335784912109, "target_z": 0.14648431671407733, "ce": 6.149696350097656, "genuineness": 4.2751336097717285, "identity": 0.36555367708206177, "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/casual_not_formal/dce0c9057eff22fc0583d20764d07e6713cfe1874c43a55d08c4f799fd179472.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": "dd07183e78c14ad62105cc625dc654eb130e5502b2dd2164321c67ab30d22418", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7b80058e52f15d17601d646f0ed5b3a9eb71aa33e4f9753da749bd79650401d4", "generation_seconds_batch": 24.068461124086753, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.114304065704346, "target_z": 0.17651371893167347, "ce": 6.109982967376709, "genuineness": 4.400654315948486, "identity": 0.7847436666488647, "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/casual_not_formal/dd07183e78c14ad62105cc625dc654eb130e5502b2dd2164321c67ab30d22418.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "dd57bc6698eb09c7624fa33725af6299846c2bd84e88b7fc981130fbbe65c5d6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bd67664b5d8aabe187ff9ac41cd2fc70434183b7ae4c33d48a3d8671d323536a", "generation_seconds_batch": 24.21990254498087, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.908714771270752, "target_z": 0.14013670077455884, "ce": 6.125868320465088, "genuineness": 4.218649387359619, "identity": 0.49255263805389404, "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/casual_not_formal/dd57bc6698eb09c7624fa33725af6299846c2bd84e88b7fc981130fbbe65c5d6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "dd6c5d63fb6ec8a8c0eceecd2738f0de5bea316e3746d7df8a2ad6cd19778eba", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "86577594e60d8ac1ecefb74c29e79db546254c6f10f202f7a68c52b2df11bc05", "generation_seconds_batch": 25.33540375402663, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6671702861785889, "target_z": -1.66015632909546, "ce": 6.292664527893066, "genuineness": 4.325342178344727, "identity": 0.4250492751598358, "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/casual_not_formal/dd6c5d63fb6ec8a8c0eceecd2738f0de5bea316e3746d7df8a2ad6cd19778eba.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "dda10fb66cb545f451224cdb25eab1d88ec44f919a62c774fe6bb3daeee2c8ca", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cfc7409412457c17172fc45996178c3817bbb4d8574768e9baf6a70452bf69e6", "generation_seconds_batch": 17.655254842014983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.134239673614502, "target_z": 0.118652247680957, "ce": 6.23706579208374, "genuineness": 4.507347106933594, "identity": 0.41239631175994873, "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/casual_not_formal/dda10fb66cb545f451224cdb25eab1d88ec44f919a62c774fe6bb3daeee2c8ca.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": "de018e1a4e95918021b005981b7ea0cbf6854138d024f23bf6f7d242cb3b0304", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The road turns left behind the bridge.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "dc9ad45667ecc1e2ef7a4e7749998b12b441f2169cbba8388c02944910362066", "generation_seconds_batch": 16.110939012025483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1773159503936768, "target_z": -0.47656248556858416, "ce": 6.332377910614014, "genuineness": 4.068024158477783, "identity": 0.20480570197105408, "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/casual_not_formal/de018e1a4e95918021b005981b7ea0cbf6854138d024f23bf6f7d242cb3b0304.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "de0d169b68a84f32ff35a176653f0025f8d547c2ac90e4b26eba5cb1d6f149c0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bbdf4c2b5bbde608c9a8c52a97607975d7b15c0d7496c16aeb08e027b0c344d8", "generation_seconds_batch": 22.268308773986064, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8035175800323486, "target_z": -0.6855468594704801, "ce": 5.6254801750183105, "genuineness": 4.143336772918701, "identity": 0.218687042593956, "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/casual_not_formal/de0d169b68a84f32ff35a176653f0025f8d547c2ac90e4b26eba5cb1d6f149c0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "de24942cc8020df1d426760de71f8e7ff850b1498fbb90556ae206c201a4103d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "314d860e53b1ed9fe29889b1f5fc07ba112b0fe95f5bc5901906f437b627b9ed", "generation_seconds_batch": 17.694343962939456, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.617330551147461, "target_z": -1.8437500305836205, "ce": 5.998785495758057, "genuineness": 4.042920112609863, "identity": 0.6229802966117859, "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/casual_not_formal/de24942cc8020df1d426760de71f8e7ff850b1498fbb90556ae206c201a4103d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "de5437f3fee327a3e42d699d8f51031be617ea57db685c0dd8975fc5ab7575e8", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c7145c9f55e47136b289cf6b303d840e55b5b4497d2e6c120946575ee2701b71", "generation_seconds_batch": 16.71694788907189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.41406245676257736, "ce": 6.205295085906982, "genuineness": 4.306513786315918, "identity": 0.3809462785720825, "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/casual_not_formal/de5437f3fee327a3e42d699d8f51031be617ea57db685c0dd8975fc5ab7575e8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "deccfa2175e9ec7bc307c254acda3dac1e5726346ca362ed06da624200df52cc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ea2376968813e94ea6158b6017954f446631cdadcef5eeb72e6f3697cdf7e7f", "generation_seconds_batch": 17.97157822700683, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2720115184783936, "target_z": -0.4960937290543983, "ce": 6.221180438995361, "genuineness": 4.507347106933594, "identity": 0.6004230976104736, "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/casual_not_formal/deccfa2175e9ec7bc307c254acda3dac1e5726346ca362ed06da624200df52cc.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": "df91e9b8f3ed113856b838f15e86c1113605a3f1e8d17c335bcafa101f1af4f6", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The keys are on the kitchen table.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b25d4825a62e2f55a94fe7556c8ffbabf5ed8096690ebee6e2250182b5468fca", "generation_seconds_batch": 21.94623415195383, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8594980239868164, "target_z": -0.03442381128127868, "ce": 6.316492557525635, "genuineness": 4.036643981933594, "identity": 0.5177619457244873, "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/casual_not_formal/df91e9b8f3ed113856b838f15e86c1113605a3f1e8d17c335bcafa101f1af4f6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "dfaefb7b78629a01c65490dbdfeaf444823ea0c209d74a3b26302aa0d602edd7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "df358c84d70539c96a37376effb79f080306c444e98a45f170baa05e69e28bb1", "generation_seconds_batch": 24.388761150068603, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7536778450012207, "target_z": -0.7519531145287981, "ce": 6.324435234069824, "genuineness": 4.080576419830322, "identity": 0.394074946641922, "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/casual_not_formal/dfaefb7b78629a01c65490dbdfeaf444823ea0c209d74a3b26302aa0d602edd7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "e01032bccf3c3eaa1a930cab00de0ea190b14a3be39947abe817eb159592c5de", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0b459a77a253d24a8aa3e0013637ce230309eb813b0bcb64bcbba8366bde5eb4", "generation_seconds_batch": 18.71418038289994, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.0585938371327441, "ce": 6.046441555023193, "genuineness": 4.143336772918701, "identity": 0.6915424466133118, "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/casual_not_formal/e01032bccf3c3eaa1a930cab00de0ea190b14a3be39947abe817eb159592c5de.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "e01d5c2755b0c44000b542412c642a7181604240fe085adeae78294f2749bfce", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "44d85ee890497fe7a36fd25654c392b034ede8eff3cacebf2384df64ea6c66d6", "generation_seconds_batch": 22.73846703604795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.987924814224243, "target_z": -0.794921923010808, "ce": 5.903473377227783, "genuineness": 4.406930446624756, "identity": 0.7501031160354614, "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/casual_not_formal/e01d5c2755b0c44000b542412c642a7181604240fe085adeae78294f2749bfce.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "e10fb06c0a29c0363d33198b4a1baf5f14d239100ec20da94c697ecf4a7d36ac", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0e2bebeb8fa192026bfd0f7284991af83b4a54c7fa6ac633cba5310cd9a7d909", "generation_seconds_batch": 16.743776149931364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.614126205444336, "target_z": -1.0742187518008275, "ce": 6.09409761428833, "genuineness": 4.507347106933594, "identity": 0.5460970401763916, "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/casual_not_formal/e10fb06c0a29c0363d33198b4a1baf5f14d239100ec20da94c697ecf4a7d36ac.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "e12e34da536bddcbda94d32b6330415e47fb00eb8f93515d1fd35587c5791830", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51efa5a0ad7f622f58c93e591d9941befa1187b21f8f15f56eee1a927b170671", "generation_seconds_batch": 26.366761818062514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.28696346282959, "target_z": -0.5527343066182525, "ce": 6.070269584655762, "genuineness": 4.256305694580078, "identity": 0.7250211238861084, "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/casual_not_formal/e12e34da536bddcbda94d32b6330415e47fb00eb8f93515d1fd35587c5791830.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "e1db7809f6bc4706ac1a5e8211d16e06c91c7cc948cdc5dc8ad95d105961a943", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2c7a34fc1760db024d951f81050a9508e7bd61fb602c65e74b37c196540d50c5", "generation_seconds_batch": 24.595484151039273, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.615906000137329, "target_z": -0.07678224619875275, "ce": 6.070269584655762, "genuineness": 4.093128204345703, "identity": 0.8151130676269531, "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/casual_not_formal/e1db7809f6bc4706ac1a5e8211d16e06c91c7cc948cdc5dc8ad95d105961a943.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e243a7b79a0dcfc0d9b4c2ed094648f4551e9d20a7d8ae85adaf661d900aa6d9", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f2833d263c60c91001e3a9a9c121bb4019878aefdb874dd643ebfa9143e4832", "generation_seconds_batch": 16.854664427926764, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.463894844055176, "target_z": -0.4990233648651007, "ce": 6.41180419921875, "genuineness": 4.444586753845215, "identity": 0.3216692805290222, "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/casual_not_formal/e243a7b79a0dcfc0d9b4c2ed094648f4551e9d20a7d8ae85adaf661d900aa6d9.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e2541a1ea8754a99aa8be2e33e0d7f629629d6b51248527b04aa3ebe4f9aa60f", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ab15abd600923cc20a656084d96eb257f9173c96de10d853d96af62f704aaa18", "generation_seconds_batch": 18.429428944014944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.053250312805176, "target_z": 0.11181644686662104, "ce": 6.205295085906982, "genuineness": 4.3504462242126465, "identity": 0.35173505544662476, "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/casual_not_formal/e2541a1ea8754a99aa8be2e33e0d7f629629d6b51248527b04aa3ebe4f9aa60f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "e30b6c716d416deb91a8c870260026e525754bfc82b75ff15e6c2edee1950163", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fc327104756209008a753a86ced3b284fa3037fddd3f5e8112fdc777aa96c510", "generation_seconds_batch": 29.448011026950553, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1349520683288574, "target_z": -0.4462891056805956, "ce": 6.09409761428833, "genuineness": 4.1998209953308105, "identity": 0.7114458680152893, "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/casual_not_formal/e30b6c716d416deb91a8c870260026e525754bfc82b75ff15e6c2edee1950163.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "e38cec167fc365e01937b0a854dc8559d9f945acbeb56fe8e79f1d998929a3f0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f09f9d021eb4e852f894b574e9c837a2676d21802eb10d4169db4b3b49685dbe", "generation_seconds_batch": 16.71694788907189, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -0.7099609069608002, "ce": 6.109982967376709, "genuineness": 4.13078498840332, "identity": 0.2544041574001312, "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/casual_not_formal/e38cec167fc365e01937b0a854dc8559d9f945acbeb56fe8e79f1d998929a3f0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e3f79d0f7cf517afa92ca83951e4f720516bb67b9f9eb317fe8b992028970538", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bc783affc45f7ea61372caceeb657b5c2ac8f7a9ec0e3aa011b863457b0cb332", "generation_seconds_batch": 18.18650393805001, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9373726844787598, "target_z": -0.004882824132532895, "ce": 6.24500846862793, "genuineness": 4.394378662109375, "identity": 0.4914018511772156, "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/casual_not_formal/e3f79d0f7cf517afa92ca83951e4f720516bb67b9f9eb317fe8b992028970538.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "e4010ab7cb655524edcf7618c006e85a0fabe5c48c8661e6bc2120efd405dfbc", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "51830656ac1220da7e239bc965842f5ab17a02285ff352158b608e6794d52f0a", "generation_seconds_batch": 16.88869151403196, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.9031970500946045, "target_z": -0.9687499378707195, "ce": 6.205295085906982, "genuineness": 4.331617832183838, "identity": 0.6790348887443542, "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/casual_not_formal/e4010ab7cb655524edcf7618c006e85a0fabe5c48c8661e6bc2120efd405dfbc.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "e4292cdf500358ecc5246dce6e5f34b4dd3cb4cdc7144c2e95850a8a476b9634", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "11acf4daeede6b7bc959f5dc9154110d98dab2937503e02027cd350991889cb6", "generation_seconds_batch": 17.96715675597079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5493345260620117, "target_z": -0.9589843997393422, "ce": 6.078212261199951, "genuineness": 4.256305694580078, "identity": 0.7851024866104126, "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/casual_not_formal/e4292cdf500358ecc5246dce6e5f34b4dd3cb4cdc7144c2e95850a8a476b9634.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "e4a372995552481ebde7fab1e09fa4f1ed02b1a06dd5039722a3aa0ba44db112", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2402adb8ed9f55448a67ce3347663734044c7d65027c90baf3edbed711bb4327", "generation_seconds_batch": 22.837030007969588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.4062499656322451, "ce": 6.300607204437256, "genuineness": 4.425758361816406, "identity": 0.46399641036987305, "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/casual_not_formal/e4a372995552481ebde7fab1e09fa4f1ed02b1a06dd5039722a3aa0ba44db112.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e529849113cd6e3d3c9e6d9b4fc9c93e6f1af78791fba38936a3bae3093925a4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8a0c6bfd80276787b2007b3302bd4abd04c86f58b5ceb35479d4b033272bf190", "generation_seconds_batch": 24.377253305050544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.79813289642334, "target_z": -0.11901851767993346, "ce": 6.197352409362793, "genuineness": 4.231201171875, "identity": 0.4854540526866913, "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/casual_not_formal/e529849113cd6e3d3c9e6d9b4fc9c93e6f1af78791fba38936a3bae3093925a4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "e53dcd60179be2890fd6f35a62abfe81ccd10c82895639c3691bbd04c676cf17", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02e8ce350c5417015d5e541689631cc9d04f18205112781d83e615e2e88bf2c0", "generation_seconds_batch": 24.91230469592847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.243887424468994, "target_z": 0.3398437510909005, "ce": 5.982900142669678, "genuineness": 4.406930446624756, "identity": 0.8236030340194702, "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/casual_not_formal/e53dcd60179be2890fd6f35a62abfe81ccd10c82895639c3691bbd04c676cf17.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "e57f0ccd64a91e36361703981a42adc64f1df3d2afdfdb189f0eb39cc1f8b741", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2df0aecd326897eee9eb989c4c893068570da98eb57a41a647cf4565b615e1f", "generation_seconds_batch": 15.362856520921923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.358163595199585, "target_z": -1.9921875790730499, "ce": 5.967014789581299, "genuineness": 4.049196243286133, "identity": 0.517910361289978, "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/casual_not_formal/e57f0ccd64a91e36361703981a42adc64f1df3d2afdfdb189f0eb39cc1f8b741.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e5a44429e526cb8ba4bf3106fb95e93db1705241d6abc9c7882ade211670e43a", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "972ad3c50e79771bf7072142a1f8266714d2cb710a9eea465c3c484dcef85153", "generation_seconds_batch": 18.717890892992727, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.233919620513916, "target_z": 0.16577154909137, "ce": 6.38797664642334, "genuineness": 4.457139015197754, "identity": 0.2995542287826538, "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/casual_not_formal/e5a44429e526cb8ba4bf3106fb95e93db1705241d6abc9c7882ade211670e43a.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": "e5a69f0c017b2c2f81c88eeef93d944f69ebcbeb588276ba471afe7035cdf9bf", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "86ace19af601be81c311a2ae57fcbc2ce28d26ba10e674743d2587b047202c78", "generation_seconds_batch": 16.3283529530745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.461402654647827, "target_z": -0.32421877698517165, "ce": 6.181467056274414, "genuineness": 4.369274139404297, "identity": 0.40996479988098145, "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/casual_not_formal/e5a69f0c017b2c2f81c88eeef93d944f69ebcbeb588276ba471afe7035cdf9bf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "e6a2b0c0882a813d71ac9700f67aec30c291a85c9a83c9b76b9e77ead47c63a4", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "de540e7d73aa0989bdaf47e5146a187b4f73befcc5fe99bab1298d9e1ec11f55", "generation_seconds_batch": 17.366269557969645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.604158401489258, "target_z": -1.1132811518654127, "ce": 6.062326908111572, "genuineness": 4.20609712600708, "identity": 0.6657631993293762, "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/casual_not_formal/e6a2b0c0882a813d71ac9700f67aec30c291a85c9a83c9b76b9e77ead47c63a4.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "b21230a81f1166931cf8b5bbde209b6f39b5e587568ffddcfc4ead436fcad801", "generation_seconds_batch": 17.1706567479996, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4215309619903564, "target_z": -0.49414061413341753, "ce": 6.340320587158203, "genuineness": 4.400654315948486, "identity": 0.4179114103317261, "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/casual_not_formal/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "e771271e43dc45f45d8d85c42800d7cafd87a994478d8c50ec005b4cb011be57", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fdd12c75ef1375cac2d16bf3721bf638f0d9c76f029be68845992b79cc21d155", "generation_seconds_batch": 21.978963394998573, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5423924922943115, "target_z": -0.20410151621698125, "ce": 5.998785495758057, "genuineness": 3.8797430992126465, "identity": 0.36377114057540894, "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/casual_not_formal/e771271e43dc45f45d8d85c42800d7cafd87a994478d8c50ec005b4cb011be57.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "e78021faf04dedf0a2761e4a8a984dd6c27f2e927f80023276d1993af8162f91", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07f19e203593d314f4770aabe65bf0feba37a84f39a9b3f09024ad323805d8a7", "generation_seconds_batch": 23.922581055085175, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5436384677886963, "target_z": -0.3037109071724959, "ce": 6.046441555023193, "genuineness": 4.469690799713135, "identity": 0.4724031090736389, "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/casual_not_formal/e78021faf04dedf0a2761e4a8a984dd6c27f2e927f80023276d1993af8162f91.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e7a1ea14ef9be8fda42ef1e13c0d4e1f6222caf2ea28a97820c81e1dfcbd5d1e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13193e51c5587b15f1f208d288b6003357b3a28c7e439220d2933fbbb18c5db5", "generation_seconds_batch": 21.32025559002068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.246379375457764, "target_z": 0.2080078040777388, "ce": 6.165581703186035, "genuineness": 4.375550270080566, "identity": 0.14670710265636444, "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/casual_not_formal/e7a1ea14ef9be8fda42ef1e13c0d4e1f6222caf2ea28a97820c81e1dfcbd5d1e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "e7c1096835c4ec74ca1e93809972f5b0d0e250f833ba504fd652192bcc6f136b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P04", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Think while speaking, with small organic hesitations and spontaneous emphasis.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "92213f0311ddc8944a184dc86ddbae96ce18f1630e4055a3787bc0fe1558c666", "generation_seconds_batch": 18.104842783068307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.187283992767334, "target_z": -0.6269530935199674, "ce": 6.006728172302246, "genuineness": 4.394378662109375, "identity": 0.7563038468360901, "hypothesis": "He encontrado la cajita. Esta junto a la ventana vieja. ¿Podemos mirar juntos lo que hay dentro?", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/e7c1096835c4ec74ca1e93809972f5b0d0e250f833ba504fd652192bcc6f136b.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": "e7e7386f1c701136fb7a0942a1158cf5b1677456f60881cb8855cf44ed962b2c", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1b566dbabda8dacb8bc5e581d83083e080b6cf2cb469b91e8c4a5074d39b0a05", "generation_seconds_batch": 15.98680782399606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.209711790084839, "target_z": -0.51660155645723, "ce": 5.935244083404541, "genuineness": 4.507347106933594, "identity": 0.8054284453392029, "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/casual_not_formal/e7e7386f1c701136fb7a0942a1158cf5b1677456f60881cb8855cf44ed962b2c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "e7e932e5f8c26b0a067eb4b9cc863734918a87e2e604f0c01e474d020ff5bf02", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "627b4736c70f640fbce3e7f4805bff8f6badedbf42f03a4fe830da3e59e73cf3", "generation_seconds_batch": 21.86924408201594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1157283782958984, "target_z": -1.4843750794604273, "ce": 6.125868320465088, "genuineness": 4.137060642242432, "identity": 0.4915758967399597, "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/casual_not_formal/e7e932e5f8c26b0a067eb4b9cc863734918a87e2e604f0c01e474d020ff5bf02.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "e7ee9a98ac468cde530886568328d03bbef94306505eea303d0fb51e6c2aaad5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fad14a494745d428549bc5412e65dc002f723cd397a0dad4e27f01d1025dde4d", "generation_seconds_batch": 23.440093860030174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8258562088012695, "target_z": -0.04272462764537506, "ce": 6.1020402908325195, "genuineness": 4.306513786315918, "identity": 0.7022531032562256, "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/casual_not_formal/e7ee9a98ac468cde530886568328d03bbef94306505eea303d0fb51e6c2aaad5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e8eabbb4de3f69827aad3774d2fe46536984f51bade6fc4203b6a6ddfc398bda", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a93de732ff1fa0463d13958f15dcbac76b9f390077506bc6d15b24c61b8e2d65", "generation_seconds_batch": 24.430546098970808, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.253855228424072, "target_z": 0.2929686825586815, "ce": 6.213237762451172, "genuineness": 4.281409740447998, "identity": 0.3554142713546753, "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/casual_not_formal/e8eabbb4de3f69827aad3774d2fe46536984f51bade6fc4203b6a6ddfc398bda.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e915d936d5578b941d789fe585f73a3aa6953967d46c435c0fccea04f89e2e96", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "02f13cafe098e1711927243be5a9cf44ac48b423091c8f31c000393d5737967c", "generation_seconds_batch": 23.290918337064795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.174111843109131, "target_z": 0.13085941235632711, "ce": 6.308549880981445, "genuineness": 4.306513786315918, "identity": 0.2479965090751648, "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/casual_not_formal/e915d936d5578b941d789fe585f73a3aa6953967d46c435c0fccea04f89e2e96.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "e9cd625915a4839be7305b2ce4f8112f35d848d264c402e9f2c158ac04e006ac", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dcc969812b2fc06111af6c26e27e2b726d579ece0a8f2298e15e861f46de5ede", "generation_seconds_batch": 17.054144437075593, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2752156257629395, "target_z": -1.1953125624766427, "ce": 6.316492557525635, "genuineness": 4.344170093536377, "identity": 0.7800543308258057, "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/casual_not_formal/e9cd625915a4839be7305b2ce4f8112f35d848d264c402e9f2c158ac04e006ac.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "e9d0200f5176d60c756b86c4722e43c8c41bc56b555b567a8af3acbab8297593", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ad0b9bce1a691bd133a3e70f475ee8deb0119eb67727d482845d2ceb094300c", "generation_seconds_batch": 22.762878520064987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2620434761047363, "target_z": -0.37597662949111166, "ce": 6.205295085906982, "genuineness": 4.118232727050781, "identity": 0.3566438853740692, "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/casual_not_formal/e9d0200f5176d60c756b86c4722e43c8c41bc56b555b567a8af3acbab8297593.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "15abe00e8e7d4eb2d0ca81b1b039bcec72254ef0ee6c31c539ad44a1e3a21f74", "generation_seconds_batch": 15.827006644918583, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.499494791030884, "target_z": -1.2246094555969478, "ce": 6.300607204437256, "genuineness": 4.557555198669434, "identity": 0.6308501958847046, "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/casual_not_formal/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "e9f5e01662a5817b5d3dcacde7012111d005468bfd84b993af8da23281e948ff", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40e64ba94818070955ba1ccb6612722ad5f9d57754c484e54a6c5f69bd094cd0", "generation_seconds_batch": 18.567688851966523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5598363876342773, "target_z": -0.2485351132606916, "ce": 5.982900142669678, "genuineness": 4.2249250411987305, "identity": 0.7310560941696167, "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/casual_not_formal/e9f5e01662a5817b5d3dcacde7012111d005468bfd84b993af8da23281e948ff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "e9feb432bac4c7ba920b3ea84d10bdd51563d792cd12f8ea8d017497ec404736", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "87fa6d270af221e5e6dbf6cfb9552c71e8b5dfec6310c83532f02b2241cea358", "generation_seconds_batch": 25.62851645902265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.3186469078063965, "target_z": 0.22729486900185916, "ce": 6.173524379730225, "genuineness": 4.293961524963379, "identity": 0.2303491234779358, "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/casual_not_formal/e9feb432bac4c7ba920b3ea84d10bdd51563d792cd12f8ea8d017497ec404736.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": "ea83176e777991a0810160c0264f5b32add08dde3e2e5fd3216d169087c6d94b", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e05c57631a2f06e23bbdbde3905cb1c5acb0cfa1025add36cbc1025cb5a52e71", "generation_seconds_batch": 16.53924127900973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.269519567489624, "target_z": -0.6025390255758729, "ce": 6.062326908111572, "genuineness": 4.243753433227539, "identity": 0.7032148838043213, "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/casual_not_formal/ea83176e777991a0810160c0264f5b32add08dde3e2e5fd3216d169087c6d94b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "ea9f0a4888c6f9022d637803b38a00152897c929a7f6bf5c8ada16a899cc4bea", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "081347751e929a51930faddc94dfd5b71181072f3ab11dbe8eaedd58742210bc", "generation_seconds_batch": 25.458952556014992, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.468878746032715, "target_z": -0.27148437057437047, "ce": 6.229123115539551, "genuineness": 4.570107460021973, "identity": 0.7135927081108093, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/ea9f0a4888c6f9022d637803b38a00152897c929a7f6bf5c8ada16a899cc4bea.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "eadebf1a8eac24bb2a1c37f9ce74d76dcfd9488a506d462093209f3360f999d6", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b26866ed6fd0c6aea986ed8684242b226fa3b975610752d60fe809113720ce51", "generation_seconds_batch": 16.635965156019665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.444671154022217, "target_z": -1.2304687316555891, "ce": 6.109982967376709, "genuineness": 4.256305694580078, "identity": 0.4346577227115631, "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/casual_not_formal/eadebf1a8eac24bb2a1c37f9ce74d76dcfd9488a506d462093209f3360f999d6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "eafea2540a6d27bbf851800d2d49aa518652cdef57338c659fa9cc0a110a2452", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2af376d9f7c1f102e9fd4bb6b208015932071490a3293e9d3ecc4f7f21ae5cce", "generation_seconds_batch": 18.567688851966523, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.847661018371582, "target_z": -0.09277346728405467, "ce": 6.125868320465088, "genuineness": 4.381826400756836, "identity": 0.6541730165481567, "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/casual_not_formal/eafea2540a6d27bbf851800d2d49aa518652cdef57338c659fa9cc0a110a2452.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "eb145f767bff91f05a12df9c5ee7aedb798e901f6c751c57276c3c64d8debb41", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7c152f7563eda6b27f1e3aaf4462d53321fddbb383408f566da968e540a3ad21", "generation_seconds_batch": 18.315314706997015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.3984375962489242, "ce": 6.022613525390625, "genuineness": 4.331617832183838, "identity": 0.7455573678016663, "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/casual_not_formal/eb145f767bff91f05a12df9c5ee7aedb798e901f6c751c57276c3c64d8debb41.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "eb64986847efe3decd837e26752285a8e828eb5e4284218632eac257f28705db", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e25e318373a8529decd7495c304a81a25321b4b05b79feef1d13762df3a0d4d8", "generation_seconds_batch": 26.07389770296868, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.499494791030884, "target_z": -1.1230469389717264, "ce": 6.348263263702393, "genuineness": 4.293961524963379, "identity": 0.6612714529037476, "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/casual_not_formal/eb64986847efe3decd837e26752285a8e828eb5e4284218632eac257f28705db.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "ebb6e81cd47428aa1e502c9fbc9b4a67ae53e79281dcea87f7f46aa1f12d871d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "272f9525c419b25cb24e5f3b07f00f626f880b7a802fba1b3a065b1569b83684", "generation_seconds_batch": 22.38128261792008, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.042036056518555, "target_z": 0.07421864332989712, "ce": 6.0146708488464355, "genuineness": 4.432034492492676, "identity": 0.611899197101593, "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/casual_not_formal/ebb6e81cd47428aa1e502c9fbc9b4a67ae53e79281dcea87f7f46aa1f12d871d.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2e89104cbd4aa4c20cee2aa49373d82920018e8f5b274a41d2a6e8abede36401", "generation_seconds_batch": 16.339050746057183, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.557344436645508, "target_z": -0.2583007132768379, "ce": 6.1020402908325195, "genuineness": 4.7332844734191895, "identity": 0.7784557938575745, "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/casual_not_formal/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "ec4d42b355bb24f0a42e289e09197c734e8babfd3241b4fd5742fb122530d72b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3bb5318d8cdf338ce0c303fdb4bc751ef9c06f90c6aa06b42f1530308bc1c936", "generation_seconds_batch": 24.420667428988963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.5443506240844727, "target_z": -0.9726561919830039, "ce": 6.23706579208374, "genuineness": 4.632867813110352, "identity": 0.5973078012466431, "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/casual_not_formal/ec4d42b355bb24f0a42e289e09197c734e8babfd3241b4fd5742fb122530d72b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "ed3c3b8dd73895e9ac9e0bd34ebe0f9f7ba3ab972e9653fe1291831f2d354c02", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6b74cbfb819fe6566bfabc3e4841dfa39a18f7335f6ecfd4e7204c7da39378e0", "generation_seconds_batch": 22.496173362014815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -1.2851562138074084, "ce": 6.427689552307129, "genuineness": 4.306513786315918, "identity": 0.5158480405807495, "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/casual_not_formal/ed3c3b8dd73895e9ac9e0bd34ebe0f9f7ba3ab972e9653fe1291831f2d354c02.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "ed5e4ab098b66eeb963b68f8650ad6513a54db219e5ee4ae4c2ffc21ae668be6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5799b06b56e97e0bdae73f1e59effecfb1f7714075e85ffd5d5cfaf46c6e394", "generation_seconds_batch": 23.707004179013893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.936126708984375, "target_z": -0.051269536047799436, "ce": 6.348263263702393, "genuineness": 4.545003414154053, "identity": 0.3352903723716736, "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/casual_not_formal/ed5e4ab098b66eeb963b68f8650ad6513a54db219e5ee4ae4c2ffc21ae668be6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "ed634e02135152f05183c98e78652970c184d8d1f0f35384ce544414d5272fbe", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "bf0f7f57ceb225c3c27df9c1fc43c76e12186bbb035246c698649c2242f96ac9", "generation_seconds_batch": 25.2426143969642, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3542470932006836, "target_z": -0.3808594315055047, "ce": 6.1894097328186035, "genuineness": 4.375550270080566, "identity": 0.33749884366989136, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Manthe 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/casual_not_formal/ed634e02135152f05183c98e78652970c184d8d1f0f35384ce544414d5272fbe.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "edefc4221d47534dd3bc630bae2233dafd711beda9483d3bda7a1843069d99df", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5b35b37372569b188b7045d2233af3e775a435249eb11a4b4b362ba615fd7869", "generation_seconds_batch": 16.408794441027567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3948311805725098, "target_z": -1.3164062424013179, "ce": 6.260893821716309, "genuineness": 4.231201171875, "identity": 0.7581822872161865, "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/casual_not_formal/edefc4221d47534dd3bc630bae2233dafd711beda9483d3bda7a1843069d99df.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "ee214b3e90f672c96c4a2904b589a580af3bf58285933caf8cbf79c78d079527", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "55a14cd7bc6bc38b3ad544f8daeaff96ae18b3ac77d1fb0178f4c5a502dc4292", "generation_seconds_batch": 16.181671860977076, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.301915407180786, "target_z": -0.3632812348570581, "ce": 6.062326908111572, "genuineness": 4.287685871124268, "identity": 0.7934890985488892, "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/casual_not_formal/ee214b3e90f672c96c4a2904b589a580af3bf58285933caf8cbf79c78d079527.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "ee2495c1da28013a32e92a3f84d109d17ef3d3f192771686da58e85c5405d8d2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a207f806c0088b43198e84bb5ce1f320df6e55c61ac0e3782380d52848f4881c", "generation_seconds_batch": 23.526450525969267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.619110345840454, "target_z": -0.8300780874932056, "ce": 6.260893821716309, "genuineness": 4.532451152801514, "identity": 0.552830696105957, "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/casual_not_formal/ee2495c1da28013a32e92a3f84d109d17ef3d3f192771686da58e85c5405d8d2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "ee6c4235217f6d4bff6506858b8a7843a003ee22530c28784d956a629eaa54b1", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4800ce62f0719e6ce947d836c46e497cb86d69f3280e58b223a9e65d334d1337", "generation_seconds_batch": 18.01441107899882, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.614126205444336, "target_z": -1.1093749620850208, "ce": 6.062326908111572, "genuineness": 4.394378662109375, "identity": 0.6046844720840454, "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/casual_not_formal/ee6c4235217f6d4bff6506858b8a7843a003ee22530c28784d956a629eaa54b1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "eedcf22f090f3c1838316b627217a3eb0c2114b032c3cd75c5684710ed0b1687", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a9619ea5a2316aa8b2435fc5b9d85fe78b7d10e6ac0f6e39872e7c79f09d6578", "generation_seconds_batch": 23.533801264013164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7369461059570312, "target_z": -1.722656234522504, "ce": 6.435632705688477, "genuineness": 4.337893962860107, "identity": 0.5022040605545044, "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/casual_not_formal/eedcf22f090f3c1838316b627217a3eb0c2114b032c3cd75c5684710ed0b1687.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "eedf703a1b201be796eb27cb47182156c142d3f5a4109e1518a1d31c4c59efab", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6790fef10fb741ef7233e95ef216cfe413cc23c4fb69ff292bac72037ddd9ad2", "generation_seconds_batch": 23.072459374088794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3350234031677246, "target_z": -1.0585938360890665, "ce": 6.165581703186035, "genuineness": 4.036643981933594, "identity": 0.24830730259418488, "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/casual_not_formal/eedf703a1b201be796eb27cb47182156c142d3f5a4109e1518a1d31c4c59efab.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "ef0fccc72106ffca1cc603d3dec914d81d6a56e62d235553fa976d650844b962", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "340a78863b69c806a835e6e34caed5736b9a604cc61d34826d778817974e047d", "generation_seconds_batch": 23.072459374088794, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.966209053993225, "target_z": -1.3085936844595554, "ce": 6.24500846862793, "genuineness": 4.256305694580078, "identity": 0.31368881464004517, "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/casual_not_formal/ef0fccc72106ffca1cc603d3dec914d81d6a56e62d235553fa976d650844b962.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f00a0668a1fe2becb07d128921296ad633c1a23ba34933129d84a442eb1f63d2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "09358afc7348b1f019f80c144592391536ff7747363cd3ecd0f12607d85c2961", "generation_seconds_batch": 26.556784225977026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.617330551147461, "target_z": -1.8984375213473865, "ce": 6.443574905395508, "genuineness": 4.118232727050781, "identity": 0.5597687363624573, "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/casual_not_formal/f00a0668a1fe2becb07d128921296ad633c1a23ba34933129d84a442eb1f63d2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "f06be7ed28535ea4bca0db325ee613c6ff3150c6bf95ea5410b103f6f7af4707", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "46243206fcaa0a12ae128488ea2af378d5b9e83dc6fdac84ddbe83126f260180", "generation_seconds_batch": 22.78421191603411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3667070865631104, "target_z": -0.3203125224158766, "ce": 5.855817794799805, "genuineness": 3.8671910762786865, "identity": 0.4654034376144409, "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/casual_not_formal/f06be7ed28535ea4bca0db325ee613c6ff3150c6bf95ea5410b103f6f7af4707.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": "f0c05f789cf4f5a85674487e01ca8377c44bb9d88ba793f7a44fbcfe1a1b6207", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "8a5b67ab90725c7a62ed9c503a9fb7fb17b1dd55e9475654aaa1fc3474e81ab2", "generation_seconds_batch": 18.08622430008836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.255280017852783, "target_z": -1.4140624401901793, "ce": 6.149696350097656, "genuineness": 4.2249250411987305, "identity": 0.5032254457473755, "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/casual_not_formal/f0c05f789cf4f5a85674487e01ca8377c44bb9d88ba793f7a44fbcfe1a1b6207.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "f0c689d87e4e9c826bb90d5f2ae519c9907e8b8f17f39ee1435612b92370ab1c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5a7037b4f55d9b8609afc81ec3aad1e8277578d10bff33745675feabb4519953", "generation_seconds_batch": 26.061494419001974, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1175081729888916, "target_z": -0.7148437876932796, "ce": 6.1020402908325195, "genuineness": 4.168440818786621, "identity": 0.6364582180976868, "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/casual_not_formal/f0c689d87e4e9c826bb90d5f2ae519c9907e8b8f17f39ee1435612b92370ab1c.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": "f116595c33883e41e08bf9b5e0696d5b0f1264e50f3a510b8f11eda93cdaddf7", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d1e16378e245fc7c792c45083e4be3727128891a1e4a249f59702f3669cb6339", "generation_seconds_batch": 18.769476873916574, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773613691329956, "target_z": -1.0078125733392196, "ce": 6.340320587158203, "genuineness": 4.406930446624756, "identity": 0.49668601155281067, "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/casual_not_formal/f116595c33883e41e08bf9b5e0696d5b0f1264e50f3a510b8f11eda93cdaddf7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "f144101bd6d8bd9ea10fae108855f01df0470e0eaec00808a9342755ba38ae2c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d31ac429d8de8b31d435cc78f2fe2e866961b83d73f7cb9b4da8286fe70faa7d", "generation_seconds_batch": 22.91692875092849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.464607000350952, "target_z": -1.1289063251640563, "ce": 6.054384231567383, "genuineness": 4.168440818786621, "identity": 0.6589881181716919, "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/casual_not_formal/f144101bd6d8bd9ea10fae108855f01df0470e0eaec00808a9342755ba38ae2c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "f17e1b2d2fa22315693f3aaa5ad2a14de228875bc131f1dbc03b2099fe060c53", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8bad7eaaf2c30091e4ea5cfca60c296d599c5e7af112cdbbe70e1a21eac8e63d", "generation_seconds_batch": 21.010590536054224, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2620434761047363, "target_z": -0.4687500344333865, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.6583050489425659, "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/casual_not_formal/f17e1b2d2fa22315693f3aaa5ad2a14de228875bc131f1dbc03b2099fe060c53.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f1837759ef8ada0600d80a60ecb1334181918c6b3ad26b9b7a0df29b432a912e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f05e67a0951da1583819907cfd4bf8d655669def4bd4abcbb0900a317f33635", "generation_seconds_batch": 23.533801264013164, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0260167121887207, "target_z": -1.621093864698982, "ce": 6.419747352600098, "genuineness": 4.306513786315918, "identity": 0.6495752334594727, "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/casual_not_formal/f1837759ef8ada0600d80a60ecb1334181918c6b3ad26b9b7a0df29b432a912e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f1946e9baa7bd880083dd548f8433c71bbfbbc925da0209ae22486d3acf55784", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bc32db90857ab1dad1093894aa8b270fdc07f693d2d739d0fb2f52b55cb8ac5", "generation_seconds_batch": 16.30936093092896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.9672765731811523, "target_z": 0.00976563596622708, "ce": 6.205295085906982, "genuineness": 4.325342178344727, "identity": 0.42995360493659973, "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/casual_not_formal/f1946e9baa7bd880083dd548f8433c71bbfbbc925da0209ae22486d3acf55784.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": "f1a97d4cd0b8e022b47a725375ec03c7bde83f7c102958c7219f0bfa0eeb215a", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e01f6474075c282fe7fcdb3ef5e2a9f1614ebec388e0df570c78d0e5587de7c9", "generation_seconds_batch": 25.921348684001714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7984442710876465, "target_z": -0.007568339100105941, "ce": 5.982900142669678, "genuineness": 4.607763767242432, "identity": 0.5648818016052246, "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/casual_not_formal/f1a97d4cd0b8e022b47a725375ec03c7bde83f7c102958c7219f0bfa0eeb215a.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "f1b867f232ae014a1fc8c974d09f1cd08dcf6c10889a2a926d546ca2ebede99b", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "616ae1737825ec1d14d6e9716f4e44ed90015d2676bb0ecd256a3cde3d704434", "generation_seconds_batch": 17.591778091038577, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.6539981365203857, "target_z": -0.7441406404927026, "ce": 6.046441555023193, "genuineness": 4.149612903594971, "identity": 0.5797695517539978, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/f1b867f232ae014a1fc8c974d09f1cd08dcf6c10889a2a926d546ca2ebede99b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "f214b097ec3a4f4ef47972a93f145db49c0018a972a6ed780bb4f0fce8a94ee8", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P01", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Make this audible from the opening words and keep every word clear.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "566c5fabf444a236f1444585b4eda4265d1e247bfe39322a759cfb1632a5ca69", "generation_seconds_batch": 17.18123829201795, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.404799222946167, "target_z": -1.1445312544468385, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.4933249354362488, "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/casual_not_formal/f214b097ec3a4f4ef47972a93f145db49c0018a972a6ed780bb4f0fce8a94ee8.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "4c150715ad69fd4185df5efb7cb5b2d679a78e9f8021e65bb704393184813a07", "generation_seconds_batch": 16.99237280804664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4745750427246094, "target_z": -1.2343750234795619, "ce": 6.078212261199951, "genuineness": 4.268857479095459, "identity": 0.6989749670028687, "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/casual_not_formal/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "f27fb6cee20490882384a2e97fd4e9eb3c9fb4ea641aa557740262d0f5f1a6fc", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cae9390f8145453eb6ad18c83f585800578f6879e79d8df67d8cffbe2be12126", "generation_seconds_batch": 23.49493841896765, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.713805913925171, "target_z": -0.718750011858236, "ce": 5.816103935241699, "genuineness": 4.21237325668335, "identity": 0.7636768221855164, "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/casual_not_formal/f27fb6cee20490882384a2e97fd4e9eb3c9fb4ea641aa557740262d0f5f1a6fc.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": "f2e43c9ac96ad38b881c3faf3accd1364611de5cdbbf6c81da598045087faa0d", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Escribí el número en el sobre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "822c9fd656b101046b53b25cb0c7b8d33e994782579ea6bda7e37d352526417e", "generation_seconds_batch": 24.710422587930225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4945108890533447, "target_z": -1.1601562441461157, "ce": 5.998785495758057, "genuineness": 4.319066047668457, "identity": 0.8172352313995361, "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/casual_not_formal/f2e43c9ac96ad38b881c3faf3accd1364611de5cdbbf6c81da598045087faa0d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "f2ea054a821607d54faf66b8139c82fbf5011693b7a42944b6f0310308fd6bc5", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "42b93772da997571701b6ca5c3282c81416538db7fa395263c2958236dc99d67", "generation_seconds_batch": 22.74002688098699, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.1054687593851025, "ce": 6.229123115539551, "genuineness": 4.080576419830322, "identity": 0.545189380645752, "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/casual_not_formal/f2ea054a821607d54faf66b8139c82fbf5011693b7a42944b6f0310308fd6bc5.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f32e65f69f0c4a41a9621a7c1a700e6c3de0a8c44d25857cd9942d677ee70760", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P06", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Begin with the quality contained, then let it become clearly audible.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "caaaa648f51acda0602d359cdbdf22aa5f3b3ea075897ffbfe5d0927dd4be51a", "generation_seconds_batch": 19.194249292020686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7967537641525269, "target_z": -1.7421874440159364, "ce": 6.4912309646606445, "genuineness": 4.293961524963379, "identity": 0.6712666749954224, "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/casual_not_formal/f32e65f69f0c4a41a9621a7c1a700e6c3de0a8c44d25857cd9942d677ee70760.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "f33eaa77f3f54a633a511fa0805b98d8b84dba1ec0f5b5999dc79603175788d2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9bea4dfdbb34642cfa7a4aaebaaf5f15bdbcfac88d4deb12f3fc812775496a6", "generation_seconds_batch": 21.95025450293906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.1640624679853246, "ce": 6.300607204437256, "genuineness": 4.356721878051758, "identity": 0.4367097318172455, "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/casual_not_formal/f33eaa77f3f54a633a511fa0805b98d8b84dba1ec0f5b5999dc79603175788d2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "f346fb26eb2fee9509ccdfe3ad0188037bce3ce8d0299b2f04b91362840c2b26", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f92455ab314bd63c336ef0e91d1893f913426530ee5d120062b92c7f105291d3", "generation_seconds_batch": 24.070664930040948, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.285183906555176, "target_z": -1.167968672902254, "ce": 6.229123115539551, "genuineness": 4.532451152801514, "identity": 0.6136741638183594, "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/casual_not_formal/f346fb26eb2fee9509ccdfe3ad0188037bce3ce8d0299b2f04b91362840c2b26.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": "f3cf5ce069174d96036451887b5223db36b7c45f8b1a81903df1ac98406b2d85", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "07e2dfc9fb840983c1bfb7d3e36e200eb216688d77af87f3663fb6c4234199e2", "generation_seconds_batch": 16.22309313202277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8035175800323486, "target_z": -0.9726561859802512, "ce": 6.229123115539551, "genuineness": 4.356721878051758, "identity": 0.6327332258224487, "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/casual_not_formal/f3cf5ce069174d96036451887b5223db36b7c45f8b1a81903df1ac98406b2d85.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f43ac1eafa68e7dcfdd460cd3a38f81ff6abf4b453251fa1696fb19b36096899", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3fe14c8d54658c5c21206293ed9e126dc79cd15e9659b90d89f4caa115cd8d4c", "generation_seconds_batch": 16.845799711998552, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3880674839019775, "target_z": -2.074218874591516, "ce": 6.443574905395508, "genuineness": 4.118232727050781, "identity": 0.6934852600097656, "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/casual_not_formal/f43ac1eafa68e7dcfdd460cd3a38f81ff6abf4b453251fa1696fb19b36096899.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f4c88b1f3cae37242d217fb6923574bee13ab6c3746654aef9504d1342e28abd", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "af7a5d56bf7b15a64b00c0c9378edd8c6028bd6b9aa681b45f86e39bfec117fe", "generation_seconds_batch": 16.702504288055934, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.8856639862060547, "target_z": -0.084716774796878, "ce": 6.2767791748046875, "genuineness": 4.432034492492676, "identity": 0.5114518404006958, "hypothesis": "I found the little box i it was beside the old window. We can look inside together.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/f4c88b1f3cae37242d217fb6923574bee13ab6c3746654aef9504d1342e28abd.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "f5be7029087c162a0168582b215a78e336b93c142af96b43496c79044d47e7a1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4d7b9f8e0a505f1d0b5c59649891ac6fac1dbe653ea87271bf49f702fbd1e065", "generation_seconds_batch": 18.56603705091402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.760441541671753, "target_z": -0.18847652212989222, "ce": 6.038498878479004, "genuineness": 4.494794845581055, "identity": 0.6679468154907227, "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/casual_not_formal/f5be7029087c162a0168582b215a78e336b93c142af96b43496c79044d47e7a1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f624bf113275070d6c1b6144f633c374da3b8d110725d7f86835c9b6ace3b5ff", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P07", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. React naturally to a silent partner and keep one consistent adult voice.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ddb3f043dd8a90e335a48d07753c51a6058ad69845b012537ea77ecafd01ae1e", "generation_seconds_batch": 17.62405568896793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6613848209381104, "target_z": -0.22827152476271545, "ce": 6.348263263702393, "genuineness": 4.262581825256348, "identity": 0.3409802317619324, "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/casual_not_formal/f624bf113275070d6c1b6144f633c374da3b8d110725d7f86835c9b6ace3b5ff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f650ed41041cd43fedab1caf9c4e16f401529a1e617d94d57db8e8b3790754ff", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fb6f9d71d9faff382580ee9bfb682b31d357c949eba1266d0a8001226413252c", "generation_seconds_batch": 25.12612471205648, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8085014820098877, "target_z": -0.8847656224932512, "ce": 6.348263263702393, "genuineness": 4.494794845581055, "identity": 0.6792755722999573, "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/casual_not_formal/f650ed41041cd43fedab1caf9c4e16f401529a1e617d94d57db8e8b3790754ff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f65580e7512a62e1d85fdc9c052ceff7f1cc5788509454da94ec68c0c50171d3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac8fe4799f9a68dffd3fca1d0feba2b13b2f5f86b26b670e4d7467736b886058", "generation_seconds_batch": 21.599056101986207, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.468166351318359, "target_z": 0.34765620927252816, "ce": 6.252951145172119, "genuineness": 4.425758361816406, "identity": 0.4834800958633423, "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/casual_not_formal/f65580e7512a62e1d85fdc9c052ceff7f1cc5788509454da94ec68c0c50171d3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f69adea77d7d50afcbaea26ff8faeb01940e207a475ac0da8320c66bf9f5aa2b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca3a487507011b449043ea08c6484e2cd36e45309d2f1a9cc145cfc9f5f88cd4", "generation_seconds_batch": 20.24067892297171, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.438262462615967, "target_z": 0.6308593331480125, "ce": 5.9431867599487305, "genuineness": 4.093128204345703, "identity": 0.3643227219581604, "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/casual_not_formal/f69adea77d7d50afcbaea26ff8faeb01940e207a475ac0da8320c66bf9f5aa2b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f6aea982613d1f82c7d7c79a56d4c441738990ba0fec52273390e7657014e134", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4643fc43e388dc6782db47c79f6d2487cc6ca90d898e8235553db16eb8c6472d", "generation_seconds_batch": 22.465436334023252, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.4375000616890796, "ce": 6.316492557525635, "genuineness": 4.243753433227539, "identity": 0.5259618759155273, "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/casual_not_formal/f6aea982613d1f82c7d7c79a56d4c441738990ba0fec52273390e7657014e134.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "f6b03d0facba5ce7a8bb83942c8adca946f0a450550db6be10e11ca4d6f4064d", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P08", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Carry the delivery through the global direction.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "aaf1243fa3f4dc3642903e4382f69a9e332791e72a51620ae533ac7799f2de10", "generation_seconds_batch": 15.516184744075872, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.045952796936035, "target_z": -1.359375006711876, "ce": 6.070269584655762, "genuineness": 4.168440818786621, "identity": 0.740519642829895, "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/casual_not_formal/f6b03d0facba5ce7a8bb83942c8adca946f0a450550db6be10e11ca4d6f4064d.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": "f6b3b2e1441a55b6d206322e49de283570645f7d530990ace9e33e26d4b33376", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 128, "seconds_generated": 10.24, "mp3_sha256": "c75591fabe1d31f6689fbf6e05b947514734c1f2fe076e29a013b3982b678c84", "generation_seconds_batch": 17.0212977839401, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6613848209381104, "target_z": -0.3005371206685781, "ce": 6.292664527893066, "genuineness": 4.124508857727051, "identity": 0.6567329168319702, "hypothesis": "Las llaves están en la mesa de la cocina. Miren el cajón y luego, junto a la ventana, llévatelas cuando te vayas.", "wer": 0.08695652173913043, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/f6b3b2e1441a55b6d206322e49de283570645f7d530990ace9e33e26d4b33376.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "f6f4ae520df877824bb04da911caa4524108563b10972c2035ff3678148284a7", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4080a6e855f041769c74396339e6e158b51199cb52a631d2573b63d7f87ff26a", "generation_seconds_batch": 15.448804730083793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.252075672149658, "target_z": -0.343749962206674, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.724041223526001, "hypothesis": "La porte est encore est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.1111111111111111, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/f6f4ae520df877824bb04da911caa4524108563b10972c2035ff3678148284a7.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": "f707503898ef7b67fd122d0546c5513a7f6ab48ea11190b4a03e163b12ac8327", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "64ee1f4335588cda5a3e6789b356845f12818241d5b9329071eb0c870d137aae", "generation_seconds_batch": 17.461518369033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.242107629776001, "target_z": -0.5507812786691431, "ce": 6.0146708488464355, "genuineness": 4.65797233581543, "identity": 0.7964892387390137, "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/casual_not_formal/f707503898ef7b67fd122d0546c5513a7f6ab48ea11190b4a03e163b12ac8327.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "2d333d34fb74eeee6eacb0e30fddbaca78f54fb9569644d73246bcb1adddd980", "generation_seconds_batch": 17.74653109605424, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.972972869873047, "target_z": -0.8671874892302619, "ce": 6.24500846862793, "genuineness": 4.607763767242432, "identity": 0.7241920232772827, "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/casual_not_formal/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f7b0048526b9768482cc8dd66c86872401affebe254dbf4a2eac7b463a555841", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c937601d8101c78624a83a83c40a01d5a8d2d700e28bab78b4df18709c6cf3a0", "generation_seconds_batch": 18.437574377050623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.2851561625246988, "ce": 6.435632705688477, "genuineness": 4.137060642242432, "identity": 0.5090156197547913, "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/casual_not_formal/f7b0048526b9768482cc8dd66c86872401affebe254dbf4a2eac7b463a555841.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f7bbf52d989d9297163d37d7870b4be6f93f09493c832244ceedf565f0d8a9a9", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8eebe7552ef2337dad21a40dcbbd162d5cb6bb94638b9c68a7e07fdb969a718", "generation_seconds_batch": 18.429428944014944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.303695201873779, "target_z": 0.2583007787018327, "ce": 6.252951145172119, "genuineness": 4.595211505889893, "identity": 0.26487988233566284, "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/casual_not_formal/f7bbf52d989d9297163d37d7870b4be6f93f09493c832244ceedf565f0d8a9a9.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": "f7ec6bfac70e1b9f4fe402a3e7ce1b8d63e02c4074d7598f06d8ed63851c653f", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "74087ea526660c7526b90b78f37b124f98937ea78cec6b588ef1bbe5fd8a6452", "generation_seconds_batch": 17.67508417600766, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7835817337036133, "target_z": -1.0273436848341246, "ce": 6.109982967376709, "genuineness": 4.256305694580078, "identity": 0.7249785661697388, "hypothesis": "Las llaves están en la mesa de la cocina. Miren 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/casual_not_formal/f7ec6bfac70e1b9f4fe402a3e7ce1b8d63e02c4074d7598f06d8ed63851c653f.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": "casual_not_formal", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": "25e70b15a5633a6a17be483668b3f0393a402f465007578ec52b8e2a7e3e5ae5", "generation_seconds_batch": 15.995051946956664, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.773613691329956, "target_z": -0.9335938020958812, "ce": 6.23706579208374, "genuineness": 4.482243061065674, "identity": 0.5366973876953125, "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/casual_not_formal/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "f887ea8335ef735386d305899343156320be16a268801197d789d9dc84ba1e95", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "ca7bae6681d3ac6a153a3133f02458a38bee84b71575c390373a162290c975c8", "generation_seconds_batch": 23.805179105023853, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.242107629776001, "target_z": -0.44726564581834405, "ce": 6.046441555023193, "genuineness": 4.1998209953308105, "identity": 0.5501906871795654, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er...", "wer": 0.16666666666666666, "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/casual_not_formal/f887ea8335ef735386d305899343156320be16a268801197d789d9dc84ba1e95.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "f8907adf34dd1fdbc8ede233d9dcf8533ecb32b35018795e8e72dec2ba8d2526", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "68df8c3564705bac4c5feb470a42a527263a97b272f88d940b4a2f5a7a57f141", "generation_seconds_batch": 22.78421191603411, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7935495376586914, "target_z": -0.7988281242100383, "ce": 5.982900142669678, "genuineness": 4.413206577301025, "identity": 0.5958020687103271, "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/casual_not_formal/f8907adf34dd1fdbc8ede233d9dcf8533ecb32b35018795e8e72dec2ba8d2526.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "f8b07f3685cbf4e965e74cb01f2dd42ac22f71988759ba2739ede36a6fcd0975", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5f5e48ebbb2c6009fe5f0d1c6591f793ba07c2f11963d27c1b21594e27623eda", "generation_seconds_batch": 25.209178582997993, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.559302568435669, "target_z": -1.068359330506003, "ce": 6.022613525390625, "genuineness": 4.293961524963379, "identity": 0.609308123588562, "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/casual_not_formal/f8b07f3685cbf4e965e74cb01f2dd42ac22f71988759ba2739ede36a6fcd0975.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "f8bbd3e22de51b07a636f20c9426b672be9b6549419bdad0ce93a0275d0bb9e0", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a4d0a40c0da19171fb0b9f0257b541b3260173ac0c54338decc9b61e43087e08", "generation_seconds_batch": 17.575322061078623, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.145632266998291, "target_z": -1.308593850964685, "ce": 6.229123115539551, "genuineness": 4.507347106933594, "identity": 0.5097659826278687, "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/casual_not_formal/f8bbd3e22de51b07a636f20c9426b672be9b6549419bdad0ce93a0275d0bb9e0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "f8f5d094a30aae2e28a71a2383864d88cbe83cc483842ea3d07ebc61bc1be3aa", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P03", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Speak with a clear intention toward one unheard listener standing nearby.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4a958c719d7c766493c5de4bcc8a6b8ce68a89bf5572b16787ebd74554304e42", "generation_seconds_batch": 18.315314706997015, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3748953342437744, "target_z": -1.1328124977205554, "ce": 6.221180438995361, "genuineness": 4.356721878051758, "identity": 0.7866209745407104, "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/casual_not_formal/f8f5d094a30aae2e28a71a2383864d88cbe83cc483842ea3d07ebc61bc1be3aa.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "f9541fe14f7b13a8d1debcab9ff4cfc7c7c49d6d1960fc6b4ef3932cb55f5d4d", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "89cd26a9a8ba927c92a855921ce86dc8222c4b5613ea7cf1de6086d9d73eca14", "generation_seconds_batch": 24.680514305015095, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1897759437561035, "target_z": -0.5830077526645693, "ce": 5.990842819213867, "genuineness": 4.6705241203308105, "identity": 0.7176553010940552, "hypothesis": "La porte est encore ouverte , j'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt !", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/f9541fe14f7b13a8d1debcab9ff4cfc7c7c49d6d1960fc6b4ef3932cb55f5d4d.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "f96146a3b912a45d1317458273ea467334fe30eb5ae5b606483c1713d17f6422", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6513e3a3df675ed1d125ee6dcef130bd83edebb723f9a5150886fe3d28215836", "generation_seconds_batch": 23.526450525969267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2072198390960693, "target_z": -0.5371093843819004, "ce": 6.181467056274414, "genuineness": 4.432034492492676, "identity": 0.7052475214004517, "hypothesis": "Die Tür ist noch offen. Ich hab 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/casual_not_formal/f96146a3b912a45d1317458273ea467334fe30eb5ae5b606483c1713d17f6422.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "f998d71a4b6ce58dc11cfde534d2153ac6d49262a8c5c16246ca902233d9b342", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bb432fa2657a6a5b7753016a4361758676b0dd45b21b8c242ec027421c6d006", "generation_seconds_batch": 23.542789327912033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.095792531967163, "target_z": -1.3437500099661688, "ce": 6.149696350097656, "genuineness": 3.760498285293579, "identity": 0.40014269948005676, "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/casual_not_formal/f998d71a4b6ce58dc11cfde534d2153ac6d49262a8c5c16246ca902233d9b342.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f9a69d89afdfae003eed78a28ae8e4399c4c928f293bc580b0fcc35e66be8ac1", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P05", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Establish the delivery, sustain it through the middle thought and release it at the end.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0ffbb8ae1317fd25434616d7689bab4581b18788921ff046eee4c826197099e3", "generation_seconds_batch": 17.82794892298989, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.6947152614593506, "target_z": -0.24255371448732693, "ce": 6.372091293334961, "genuineness": 4.375550270080566, "identity": 0.3955519199371338, "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/casual_not_formal/f9a69d89afdfae003eed78a28ae8e4399c4c928f293bc580b0fcc35e66be8ac1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "f9b68fb96e52e473892cb98c36f2f3019171000e7ac76f35ea7711ae5bcdc172", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "dd965a42384e6fc6dc2c8531382c317920b5fc8a4459fe7d609032dc1b2798e1", "generation_seconds_batch": 25.33752108097542, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4945108890533447, "target_z": -1.1874999895279987, "ce": 6.284721851348877, "genuineness": 4.268857479095459, "identity": 0.5492324829101562, "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/casual_not_formal/f9b68fb96e52e473892cb98c36f2f3019171000e7ac76f35ea7711ae5bcdc172.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "f9d065fea9aada42538de410307a8f232e0ff8bac484cc6ad4f3cf7aca2606a7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "83e8f6ac801220aba20a115aa99a8c0d6f0670ee6abf78117c83ab9f4ab660e8", "generation_seconds_batch": 21.740796232013963, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.7803773880004883, "target_z": -0.10839845323161146, "ce": 6.316492557525635, "genuineness": 4.356721878051758, "identity": 0.3735770583152771, "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/casual_not_formal/f9d065fea9aada42538de410307a8f232e0ff8bac484cc6ad4f3cf7aca2606a7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "fa2710ee502eb2bd63c835c4831f52fd1a48c0fcedcdcd1d930a76f257b48271", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ff00470861df323d49b8b7a7ce495e5df560c608b136489dbef1414acb44a8f1", "generation_seconds_batch": 18.830436743912287, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5860021114349365, "target_z": -0.2539062605720478, "ce": 6.141753673553467, "genuineness": 4.124508857727051, "identity": 0.6383126974105835, "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/casual_not_formal/fa2710ee502eb2bd63c835c4831f52fd1a48c0fcedcdcd1d930a76f257b48271.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "fa2dffa4050b0bc229e47605948f08ca87abf1ac8ecaf54685043e45ef492481", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "21fdd8710ca33e4241fe76b9ab20e59d36a4d527d467b6c38552c94a8c5ad10d", "generation_seconds_batch": 20.420893021975644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7486937046051025, "target_z": -0.6816406478328678, "ce": 6.1894097328186035, "genuineness": 4.005263805389404, "identity": 0.42928263545036316, "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/casual_not_formal/fa2dffa4050b0bc229e47605948f08ca87abf1ac8ecaf54685043e45ef492481.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "fa56442ecab58b9b25d0bb5fd57e528e791322b785418cd2c76c759eba0391b0", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6095508b6f2f14c8ab16f504d5e9ca3adcb952a6cda3318478fb60e63bf62d60", "generation_seconds_batch": 22.964893874013796, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.433278560638428, "target_z": 0.46386717829778096, "ce": 6.165581703186035, "genuineness": 4.582659721374512, "identity": 0.2749848961830139, "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/casual_not_formal/fa56442ecab58b9b25d0bb5fd57e528e791322b785418cd2c76c759eba0391b0.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "faaf7ba5dce4f3d8ef0b296efe14f7ee1700a55f47cec452fb385a712a83f1d2", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "03b827d620ba91e3b56dc27da34504e3c23234061a5b3437d1f9b646f60652c0", "generation_seconds_batch": 21.952278492040932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0626845359802246, "target_z": -0.5585937076862008, "ce": 5.974957466125488, "genuineness": 4.118232727050781, "identity": 0.5383790135383606, "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/casual_not_formal/faaf7ba5dce4f3d8ef0b296efe14f7ee1700a55f47cec452fb385a712a83f1d2.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "fb3833ab6c93a6e8172a1259325289e30e0492963d0f4f834ddb9831d27a0f41", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b9ca84ecfd36fad9d1001255fb0547bdef3cc57765b9131d42159ddf080c05a8", "generation_seconds_batch": 22.776833641924895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7038378715515137, "target_z": -1.011718737805611, "ce": 6.157639026641846, "genuineness": 3.9550557136535645, "identity": 0.4920526444911957, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/fb3833ab6c93a6e8172a1259325289e30e0492963d0f4f834ddb9831d27a0f41.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "fb632c7b782ec86776dba77e26b31ceb951ead2f694848b34341cb8abb7f61bf", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06968afef5898a2cf496691e82e4744adb879a03077ac2fb907f228025d22d27", "generation_seconds_batch": 18.664490289986134, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3766751289367676, "target_z": -0.425781220731382, "ce": 5.808161735534668, "genuineness": 4.293961524963379, "identity": 0.7438154220581055, "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/casual_not_formal/fb632c7b782ec86776dba77e26b31ceb951ead2f694848b34341cb8abb7f61bf.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "fc138822c7a6a65e43a6bf2a24a1669ce37c43768dd852c57bc6c497435b9c72", "bucket": "casual_not_formal", "phase": "wording", "candidate": "canonical-extension-P02", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Use breath, pitch movement, consonant attack and release to realize the requested quality.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b5c25fd2bf3bb4baba0c782d88b1a26ef6d59d85d10ac8cf472334aba957987e", "generation_seconds_batch": 16.54903685906902, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.454638957977295, "target_z": -1.2148438445217313, "ce": 6.157639026641846, "genuineness": 4.6203155517578125, "identity": 0.7219192981719971, "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/casual_not_formal/fc138822c7a6a65e43a6bf2a24a1669ce37c43768dd852c57bc6c497435b9c72.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "fc1778c27b22f3ce25a5fe611ef76b356489d6e24c6b58ea1aa657b7883e367b", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "255e4fa8fab8dec23ea855d9f58089190cd05314274ad41eb1992dc467fdff97", "generation_seconds_batch": 22.91692875092849, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.0577003955841064, "target_z": -0.4824218923135376, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.5580074787139893, "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/casual_not_formal/fc1778c27b22f3ce25a5fe611ef76b356489d6e24c6b58ea1aa657b7883e367b.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "fc4baea85ef82c0312459821162842cca0dfe324bd0cbd60d235ed2f18716dd7", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cae3b5211daa6a2fd86dc0d6e5aa11284615d155557366accc00b7fc3f9ab794", "generation_seconds_batch": 17.827782398089767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.7088217735290527, "target_z": -0.9941407131107942, "ce": 6.427689552307129, "genuineness": 4.469690799713135, "identity": 0.6856225728988647, "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/casual_not_formal/fc4baea85ef82c0312459821162842cca0dfe324bd0cbd60d235ed2f18716dd7.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-0", "key": "fc654f0f32a2ad2fa6c465273c2ea410047bbffda37dbf17e69212f1bf5bf8f6", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] The door is still open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I put your coat on the chair.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ef6c70bc2a4561902487eb61e3c169857cd6bd7a48319af08240e9c388a0c4f2", "generation_seconds_batch": 25.03010569489561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.8533573150634766, "target_z": -0.6542968738082536, "ce": 6.284721851348877, "genuineness": 4.293961524963379, "identity": 0.2868950664997101, "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/casual_not_formal/fc654f0f32a2ad2fa6c465273c2ea410047bbffda37dbf17e69212f1bf5bf8f6.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": "fd32798718d495f4ce94efb30bba90d334a1ff17de3a9d21e65b7cd16f0f68c6", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "75e3d1c582d395950e1225b642313a94177d75aee440618de54e93b5e4e70721", "generation_seconds_batch": 16.803563880966976, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.2445995807647705, "target_z": -0.6826172256823047, "ce": 6.125868320465088, "genuineness": 4.65797233581543, "identity": 0.7840371131896973, "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/casual_not_formal/fd32798718d495f4ce94efb30bba90d334a1ff17de3a9d21e65b7cd16f0f68c6.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-0", "key": "fd9f667eb1c5dc5990587e6fb43bd198f637592f05c8d37dc91a7ae90e7f3611", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d5447b46a30c08687080290582f57f9c320e98c400e6a49b7cdfcd333e1c2a3f", "generation_seconds_batch": 19.171465184073895, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.5847561359405518, "target_z": -0.2133789162217723, "ce": 5.911416053771973, "genuineness": 4.300237655639648, "identity": 0.5876172780990601, "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/casual_not_formal/fd9f667eb1c5dc5990587e6fb43bd198f637592f05c8d37dc91a7ae90e7f3611.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": "fddd194aa7b6571a2392ad577308f4fa171f4e17885d7f4918232815c7c90dc4", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Your message arrived after I had left.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "da8c8331e217f9cc9d16244c92806a4fc6edc3133c005363022e1e9ca751a599", "generation_seconds_batch": 21.793964514043182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4913065433502197, "target_z": -0.2675781111448494, "ce": 6.348263263702393, "genuineness": 4.344170093536377, "identity": 0.3701004385948181, "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/casual_not_formal/fddd194aa7b6571a2392ad577308f4fa171f4e17885d7f4918232815c7c90dc4.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "fdee180a6b55ecb4504250c93ab780d3d32c434dcd2108755371784f76217517", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "3c96d7c6db5d834436deeb2bcb89c848bc81d1db241c01625449074fdc28c53d", "generation_seconds_batch": 23.368959016981535, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.411562919616699, "target_z": -0.4003906131713778, "ce": 6.125868320465088, "genuineness": 4.607763767242432, "identity": 0.7117452621459961, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/fdee180a6b55ecb4504250c93ab780d3d32c434dcd2108755371784f76217517.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": "fdfb7896e9b911dc416587b594f32ad507a7b05d4fb833b42a65aa5d3d8df80a", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "5a7887c6408422c512a31d76a031624d52e22a7057ca63e2ffd8e880cd849f6b", "generation_seconds_batch": 17.34960559103638, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.593478202819824, "target_z": -0.307617190906361, "ce": 5.982900142669678, "genuineness": 4.482243061065674, "identity": 0.7501741051673889, "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/casual_not_formal/fdfb7896e9b911dc416587b594f32ad507a7b05d4fb833b42a65aa5d3d8df80a.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": "fe2ff1799a45295633f659ef66c7edea056fd1c8494126834d956e2f857e0ad3", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "c479cb244fe5a326fd3e2d3b7a4d872171a000abcba98a4700d7bc9f9a6d423d", "generation_seconds_batch": 24.465385763091035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4389748573303223, "target_z": -0.3740234154955122, "ce": 6.229123115539551, "genuineness": 4.18099308013916, "identity": 0.7432050108909607, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las het twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/fe2ff1799a45295633f659ef66c7edea056fd1c8494126834d956e2f857e0ad3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-es-1", "key": "fe51c07d8f25ec9ce544e48b67d7fbb86bb0c9b90de51e75aa12ad526c13343e", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2030c9a11db95b43609c1590c8cbe5223e1b1c97f5c22e8ca0bc1c28cb7edd87", "generation_seconds_batch": 16.321908641955815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.3848633766174316, "target_z": -1.4023436683273585, "ce": 6.1020402908325195, "genuineness": 4.256305694580078, "identity": 0.5110474824905396, "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/casual_not_formal/fe51c07d8f25ec9ce544e48b67d7fbb86bb0c9b90de51e75aa12ad526c13343e.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-en-1", "key": "fe83abad164182a699c688bc3a3d5a531176ca72ceeed6fffafaac68667fa2ab", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] I found the little box.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] It was beside the old window.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "401a42558b921701bbd6e85fddd76495983b328d9ecc1f79cb8244a48b8fe0fa", "generation_seconds_batch": 20.699485231074505, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.498070240020752, "target_z": 0.336425775936563, "ce": 6.292664527893066, "genuineness": 4.532451152801514, "identity": 0.4125010371208191, "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/casual_not_formal/fe83abad164182a699c688bc3a3d5a531176ca72ceeed6fffafaac68667fa2ab.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-0", "key": "fe96d839f3ebc5ab63db6bf7ce6b316d2e5a8768b187f0d5bd5689c5b03039b3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Die Tür ist noch offen.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f66576b4734b9b121d62e0db58342f9d738c3e2c2c2889113e3956dd21e10245", "generation_seconds_batch": 18.651032480993308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1673479080200195, "target_z": -0.4648437798640914, "ce": 6.165581703186035, "genuineness": 4.607763767242432, "identity": 0.4350341558456421, "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/casual_not_formal/fe96d839f3ebc5ab63db6bf7ce6b316d2e5a8768b187f0d5bd5689c5b03039b3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-1", "key": "fe9ac2ca0c77d9efc3bc9b6f50b26bd8a6afddb31434ecdb7bea93cc7743aeff", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "06bb4017df8f6594c9b54997815ae8e32392ee8363f67550baea4f9d1d992af5", "generation_seconds_batch": 17.578236727043986, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.354959487915039, "target_z": -1.3203124910326882, "ce": 5.959072113037109, "genuineness": 4.099404335021973, "identity": 0.5988130569458008, "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/casual_not_formal/fe9ac2ca0c77d9efc3bc9b6f50b26bd8a6afddb31434ecdb7bea93cc7743aeff.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-0", "key": "feaeddc79ac6bb3c719e582e625235be73e9ec56577da573699c3ec9b09a69ec", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] De deur staat nog open.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7afcf6247546d908d2fb6eca63fe4717199e5f1b0afcfbac1301367310fdde90", "generation_seconds_batch": 17.55643746908754, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.104335784912109, "target_z": 0.15771474978912317, "ce": 5.990842819213867, "genuineness": 4.287685871124268, "identity": 0.6101639270782471, "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/casual_not_formal/feaeddc79ac6bb3c719e582e625235be73e9ec56577da573699c3ec9b09a69ec.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "febcbdd17bab4e0f4ea2a3c3391210e59a69246a1685351c5fa32548c97076c3", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "1d7607b0c84cff19b52618cd4892f752439f49ca0a71036e62d15561e526e5ec", "generation_seconds_batch": 23.399708233075216, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.973506450653076, "target_z": -0.08642580815193393, "ce": 6.348263263702393, "genuineness": 4.507347106933594, "identity": 0.7305224537849426, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi, quand tu es prêt?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/casual_not_formal/febcbdd17bab4e0f4ea2a3c3391210e59a69246a1685351c5fa32548c97076c3.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-nl-1", "key": "fee7e55187f495605469e61a492b40d3fa243e7c005d20e4f723a64b3cbadb6c", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Het stond naast het oude raam.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cae9318af3a9df02ff1ced4b0efeeba86181595222f7a313a5fe95dd0d3a82eb", "generation_seconds_batch": 21.182026260066777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2353439331054688, "target_z": -1.1757811799534892, "ce": 6.173524379730225, "genuineness": 4.369274139404297, "identity": 0.6364728212356567, "hypothesis": "Ik heb het een kleine doosje gevonden. 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/casual_not_formal/fee7e55187f495605469e61a492b40d3fa243e7c005d20e4f723a64b3cbadb6c.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-fr-0", "key": "ff06adaf85ffeb5655101e07d8109bef5606fcccfb178f0928618057d2f55f78", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] La porte est encore ouverte.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b399615e578b31e1fbec302d44bd83af9b71b2b42474c0c7d9fd238b02c16592", "generation_seconds_batch": 27.279324246919714, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.3318192958831787, "target_z": -0.4882812646739265, "ce": 6.23706579208374, "genuineness": 4.6956281661987305, "identity": 0.7353397011756897, "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/casual_not_formal/ff06adaf85ffeb5655101e07d8109bef5606fcccfb178f0928618057d2f55f78.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": "ffdd66d365b09d483ade20466213b336b5f15f2ee4a782ff752921d476846ea1", "bucket": "casual_not_formal", "phase": "heldout", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction 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": "b05d1cdcc09bea71ac04c1b3d9c2fca88982d23fe23897fdd2ed0673b4ff6248", "generation_seconds_batch": 15.7392909869086, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 4.037052154541016, "target_z": -0.05761724212424282, "ce": 5.95112943649292, "genuineness": 4.645420074462891, "identity": 0.6049320697784424, "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/casual_not_formal/ffdd66d365b09d483ade20466213b336b5f15f2ee4a782ff752921d476846ea1.mp3" }, { "id": "casual_not_formal-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": "casual_not_formal-de-1", "key": "ffdfccc0897c436e6d0d2004462ceada6d23129b29c0ba0f3c620d8f97d0e5d1", "bucket": "casual_not_formal", "phase": "grid", "candidate": "canonical-extension-P09", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Very informal, loose conversational delivery to a familiar listener, with casual rhythm and spontaneous emphasis. Follow the local performance direction beside each sentence.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.5, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d78a0de0ef0f11c4768a18651f5b8b3de3007951fa1a87c945d07a377a1306b", "generation_seconds_batch": 22.837030007969588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6372663974761963, "target_z": -1.859374952298134, "ce": 6.395919322967529, "genuineness": 4.21237325668335, "identity": 0.6059706211090088, "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/casual_not_formal/ffdfccc0897c436e6d0d2004462ceada6d23129b29c0ba0f3c620d8f97d0e5d1.mp3" }, { "key": "original-casual_not_formal-de-0-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "39949d93e6d2eda4b59d1598dbd75c91359c02588b9ed16bcc621386fa7662cb", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.085824489593506, "target_z": -1.2421875441018186, "ce": 6.149696350097656, "genuineness": 4.532451152801514, "identity": 0.8064961433410645, "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/casual_not_formal/de/bd59202b736dfd0f32c23fea.mp3" }, { "key": "original-casual_not_formal-de-0-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "c0b24b597d64f46dbc2f78b6bba99ed8a35f9638b4f92e57eb18b2c069ff1430", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.095792531967163, "target_z": -1.3437500099661688, "ce": 6.213237762451172, "genuineness": 4.444586753845215, "identity": 0.510587215423584, "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/casual_not_formal/de/7ce32a171ccc5d61e6d29e0f.mp3" }, { "key": "original-casual_not_formal-de-0-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "47a8630fe1de5280adbdf5d2db4a0ecb36a915474a63cfde75d0a82fd7598ad5", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.856561541557312, "target_z": -1.546874994413439, "ce": 6.284721851348877, "genuineness": 4.394378662109375, "identity": 0.4434346556663513, "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/casual_not_formal/de/c43d519acd3dbd46cfa11d28.mp3" }, { "key": "original-casual_not_formal-de-0-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "ef093e8bcf35a16a29d5efc623f4ee9e6927c365ab02302b159c49085a1415b0", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.3948311805725098, "target_z": -1.1640625856872822, "ce": 6.213237762451172, "genuineness": 4.557555198669434, "identity": 0.4956408143043518, "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/casual_not_formal/de/71eca4ca984f58440c91b5cd.mp3" }, { "key": "original-casual_not_formal-de-0-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "72e6d07d145a5e46a43b7302ee8a86593e309645e100af835bc18cc522a7300a", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.1157283782958984, "target_z": -1.2421874852508399, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.6215035319328308, "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/casual_not_formal/de/6152f54ed1c35cd23ac898ec.mp3" }, { "key": "original-casual_not_formal-de-0-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "9b6c357056a6885b77671d8262bd0139db58434776a1c05086115418f5d97dc1", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3382277488708496, "target_z": -1.8437500446764807, "ce": 6.23706579208374, "genuineness": 4.444586753845215, "identity": 0.559662938117981, "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/casual_not_formal/de/aa4de3108b9c3d85d4dfeb3e.mp3" }, { "key": "original-casual_not_formal-de-0-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "20fa79b19be7a88063eadbd482ec21c54689288e56fdffd07dca1461e10ecf89", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.717010259628296, "target_z": -1.5390624898417415, "ce": 5.95112943649292, "genuineness": 4.3504462242126465, "identity": 0.6392016410827637, "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/casual_not_formal/de/a657154e386f5bd7ef063051.mp3" }, { "key": "original-casual_not_formal-de-0-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "cb06d66d4182c1cce4fac38ed2e1ea89872293ca3b78edafb7140202653fe848", "task_id": "casual_not_formal-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.1655683517456055, "target_z": -1.2656249560326427, "ce": 6.197352409362793, "genuineness": 4.582659721374512, "identity": 0.6372396945953369, "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/casual_not_formal/de/ccf02156206380524965abbd.mp3" }, { "key": "original-casual_not_formal-de-1-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "3515f1e2675aeadb1e9fb9a39c87013ab55fe45cf07eec09d42f3ae6d9e9d159", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.5575227737426758, "target_z": -1.8281249820333645, "ce": 6.316492557525635, "genuineness": 4.111956596374512, "identity": 0.47446319460868835, "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/casual_not_formal/de/8730fc63aa9f66477f2c91be.mp3" }, { "key": "original-casual_not_formal-de-1-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "d9a1ccf9b11714e02204692fad629bc566fd3530b56204f8e4578f763d9ab6f4", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.9064013957977295, "target_z": -1.7070312388204743, "ce": 6.459460258483887, "genuineness": 4.268857479095459, "identity": 0.5401617884635925, "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/casual_not_formal/de/521b512e9e8a8e971d1fcf00.mp3" }, { "key": "original-casual_not_formal-de-1-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "32133d0c410dc7a77ee05021c225c69c8d131072bf7bc2cb8fc05c456ec4ea1d", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.198676347732544, "target_z": -2.10937504063943, "ce": 6.459460258483887, "genuineness": 4.325342178344727, "identity": 0.450244665145874, "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/casual_not_formal/de/473df74b69664417eada60b5.mp3" }, { "key": "original-casual_not_formal-de-1-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 1.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "f70b3756816bc79eda3412b24f249a2533080dac6a9e07382320bd621a60b231", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.717010259628296, "target_z": -1.7656249177991157, "ce": 6.483288764953613, "genuineness": 4.432034492492676, "identity": 0.4694516658782959, "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/audio/casual_not_formal/de/d82d3c2863406e52e6f9a6a9.mp3" }, { "key": "original-casual_not_formal-de-1-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "fa4e833a9faf67e9432af8eeb8956a76bdd32f13f3c4da1a80e4a7777ad39bc4", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.6871063709259033, "target_z": -1.761718744198939, "ce": 6.38797664642334, "genuineness": 4.13078498840332, "identity": 0.4359918534755707, "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/casual_not_formal/de/aeb64be03d3cac23e08d3dbf.mp3" }, { "key": "original-casual_not_formal-de-1-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "18a76f8b9b3eb666a9c991bb11538d6615ecbe53e070577a8adb8b2ea208e133", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.98614501953125, "target_z": -1.5351562940585972, "ce": 6.3641486167907715, "genuineness": 4.369274139404297, "identity": 0.5105352401733398, "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/casual_not_formal/de/0f562cd9516e53c98e34cbd5.mp3" }, { "key": "original-casual_not_formal-de-1-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "a1f475188b8ac6427511f257cafc975e88c6cee200864c45ac11e463fe456df3", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.7768179178237915, "target_z": -1.6171875407738912, "ce": 6.372091293334961, "genuineness": 4.2374773025512695, "identity": 0.49779200553894043, "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/audio/casual_not_formal/de/ba04296245b336d98acaa138.mp3" }, { "key": "original-casual_not_formal-de-1-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "aa67e63636f65be9d1a348658a937b5de9005f333f29376395d89b4a36ca4f05", "task_id": "casual_not_formal-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.0559206008911133, "target_z": -1.5390624514785585, "ce": 6.332377910614014, "genuineness": 4.2249250411987305, "identity": 0.3384665250778198, "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/casual_not_formal/de/c24347fd8a657f4f37677909.mp3" }, { "key": "original-casual_not_formal-en-0-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "0012ea21a5b1a262ab6b8ff4187d025da0d02f127f02502d0d0d72bd30b1222b", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 0.8996376991271973, "target_z": -2.0078125461971768, "ce": 6.133810997009277, "genuineness": 3.7416701316833496, "identity": 0.47198933362960815, "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/casual_not_formal/en/f802ff5bf8a086207728c84e.mp3" }, { "key": "original-casual_not_formal-en-0-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "bfd945918bde62f697155e93d2292a2de2d116058acfd87dda4e9b961d7731c0", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.1755361557006836, "target_z": -1.121093806891681, "ce": 6.133810997009277, "genuineness": 4.20609712600708, "identity": 0.1574057936668396, "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/casual_not_formal/en/6971ae07ebf35612c379c1ef.mp3" }, { "key": "original-casual_not_formal-en-0-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "c077d2a05b3e4f0a852af91572f7af5374488b0ea7f35943049fbc074a71611b", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.0160489082336426, "target_z": -1.3710937629163205, "ce": 6.221180438995361, "genuineness": 3.9927117824554443, "identity": 0.3005743622779846, "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/casual_not_formal/en/ec23959156bc973ff4f82eca.mp3" }, { "key": "original-casual_not_formal-en-0-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "6038e1dadcdc81d647bcb59b5353ddfe2e3eb5781a6e1c231a59e1dfd4d6540f", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9363051652908325, "target_z": -1.3710938172004068, "ce": 6.1894097328186035, "genuineness": 3.8797430992126465, "identity": 0.41826263070106506, "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/casual_not_formal/en/82e2dca8b94ef2375194036c.mp3" }, { "key": "original-casual_not_formal-en-0-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "1d9a64698bd1eb10fcc3755b862a8755111aecf1282828d4f50bc9cc76a2aa18", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.1556003093719482, "target_z": -1.1757813057455544, "ce": 6.2767791748046875, "genuineness": 4.18099308013916, "identity": 0.13801956176757812, "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/casual_not_formal/en/4bdd16888daaa55842c8cfb8.mp3" }, { "key": "original-casual_not_formal-en-0-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "1d35b15901d532b511807819eb0001ba176584643b3acb18b912cd647fa4f34e", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9363051652908325, "target_z": -1.3085937433105341, "ce": 6.125868320465088, "genuineness": 4.030367851257324, "identity": 0.3111140727996826, "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/casual_not_formal/en/1804b828d29c8004910b539a.mp3" }, { "key": "original-casual_not_formal-en-0-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "6c9aad92fab9337b8a2a7e8965f393ca51bc6d2533e0d07a234af9106ae26dc1", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6970741748809814, "target_z": -1.5546875211681046, "ce": 6.086154937744141, "genuineness": 3.6789097785949707, "identity": 0.5283510684967041, "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/casual_not_formal/en/74075d2a631f40c9eb0de303.mp3" }, { "key": "original-casual_not_formal-en-0-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "39bd7413824c0ba0e15f7e6fc9472ad7db8e498afba4cf5697407962d22008ea", "task_id": "casual_not_formal-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Tell me when you are ready.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8964333534240723, "target_z": -1.4414063006648261, "ce": 6.24500846862793, "genuineness": 3.898571252822876, "identity": 0.21805410087108612, "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/casual_not_formal/en/1c1ea14995ac918032e94371.mp3" }, { "key": "original-casual_not_formal-en-1-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "a725c3553dc21a6d41bbfe5209dbfd6c92326fe8553eb20858eeb16b48eab70a", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.461402654647827, "target_z": -0.39453125666545674, "ce": 6.2767791748046875, "genuineness": 4.432034492492676, "identity": 0.3227882981300354, "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/casual_not_formal/en/52d3fc999662b4daf3db0b1b.mp3" }, { "key": "original-casual_not_formal-en-1-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "deaf9470370f949afc86055c75a1a02ec70e31dd09fb4f2f4a7c23c3936495fa", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.6722872257232666, "target_z": -0.25231935691331736, "ce": 6.356205940246582, "genuineness": 4.507347106933594, "identity": 0.4744432866573334, "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/casual_not_formal/en/35a87e2da3fcb8effff6ee3e.mp3" }, { "key": "original-casual_not_formal-en-1-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "df9d0111352048c87f3be54f274611960bf38161e07268f865c2fdd9d5388b18", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.756080389022827, "target_z": -0.1203613581885831, "ce": 6.300607204437256, "genuineness": 4.494794845581055, "identity": 0.454578697681427, "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/casual_not_formal/en/21f848e674feeae67ed8e8e3.mp3" }, { "key": "original-casual_not_formal-en-1-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "1474fd62a5b5156f794669665d844a789a35a3eaf9a1ae764d2f89b98b14cbcb", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.341787338256836, "target_z": -0.4570312360800173, "ce": 6.316492557525635, "genuineness": 4.532451152801514, "identity": 0.41175907850265503, "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/casual_not_formal/en/6c55b950490df4ce7f434c4b.mp3" }, { "key": "original-casual_not_formal-en-1-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "9ecd8a6a67cb641c8cae850154f2857d87b2a189a1d35a864dc3330dc1b19e54", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.137444019317627, "target_z": -0.5156250379157827, "ce": 6.395919322967529, "genuineness": 4.507347106933594, "identity": 0.32493114471435547, "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/casual_not_formal/en/a9b93a0f32fc006d0107fe95.mp3" }, { "key": "original-casual_not_formal-en-1-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "729fcf3c152534a895c9585f1c0eb6c900cbf265b4f3aaac43c1c3b6de7bef55", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.770409345626831, "target_z": -0.1445312428335158, "ce": 6.292664527893066, "genuineness": 4.469690799713135, "identity": 0.5469875335693359, "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/casual_not_formal/en/1e36cd08feb1ab2e06cdb5d0.mp3" }, { "key": "original-casual_not_formal-en-1-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "751c16ddcf7f3786c39c42c0991d8cedd12c2edc5c2536d066fcb41468254d52", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.296931505203247, "target_z": -0.48828124307762544, "ce": 6.24500846862793, "genuineness": 4.413206577301025, "identity": 0.2035089135169983, "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/casual_not_formal/en/62854342602b0256a4bfba3a.mp3" }, { "key": "original-casual_not_formal-en-1-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "d94e6a5affeab3a25e0d4a9b733ed69f2ef29f2bb663b129f32225758c8b4e15", "task_id": "casual_not_formal-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.6277430057525635, "target_z": -0.2746581961934227, "ce": 6.284721851348877, "genuineness": 4.400654315948486, "identity": 0.29068225622177124, "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/casual_not_formal/en/c64e62a46b5d81f59c245551.mp3" }, { "key": "original-casual_not_formal-es-0-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "679e81f7a6029bfab3b5d9aacbc436bef5fca1a7a533447caf07cadac5aaa735", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.197251796722412, "target_z": -0.4609374885619574, "ce": 6.09409761428833, "genuineness": 4.632867813110352, "identity": 0.7168940901756287, "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/casual_not_formal/es/44f312ef6f3546fb4633dca8.mp3" }, { "key": "original-casual_not_formal-es-0-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "8d5316f0f6ca8ab2534e15992fd4eff097906f30784857ff848b627af12dcfc9", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.987924814224243, "target_z": -0.5488280963500651, "ce": 5.863759994506836, "genuineness": 4.570107460021973, "identity": 0.6374789476394653, "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/casual_not_formal/es/fd2a8374eb220988c18c8f53.mp3" }, { "key": "original-casual_not_formal-es-0-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "d39684b3474af5b0697392c17e60577ce6bbb89d34bae756ad7c189794f0e576", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.5742545127868652, "target_z": -1.1093749782652362, "ce": 6.054384231567383, "genuineness": 4.331617832183838, "identity": 0.5338709354400635, "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/casual_not_formal/es/4e39fe2710ddbb8e9eb468cd.mp3" }, { "key": "original-casual_not_formal-es-0-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "a1ea9a9b06bad975cbfb259334054ca92e0c25885c3e6f7bbb63e891249491ce", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.8433892726898193, "target_z": -0.7265624613630893, "ce": 6.046441555023193, "genuineness": 4.425758361816406, "identity": 0.7094548940658569, "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/casual_not_formal/es/98131960ab1a97a9fadbbbe0.mp3" }, { "key": "original-casual_not_formal-es-0-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "d90b08dff8caf50dbe99272d3fcaca3c2c42e1c8b72cfada7212306ea4c36927", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.9580209255218506, "target_z": -0.5917968303875478, "ce": 5.534139156341553, "genuineness": 4.494794845581055, "identity": 0.833525538444519, "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/casual_not_formal/es/c4bde14c389cc57655e15cd8.mp3" }, { "key": "original-casual_not_formal-es-0-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "d8d26539a788c406db037f4a6988c0db205be43f177d041fab9647990a0aa7d2", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.617330551147461, "target_z": -1.8320312150063585, "ce": 6.1020402908325195, "genuineness": 3.886019229888916, "identity": 0.5768418908119202, "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/casual_not_formal/es/f769bd1394a1f57baf984e3c.mp3" }, { "key": "original-casual_not_formal-es-0-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "31305823fdeab68bda6ee88f421dd324324b1e1d11956a6f348ad1e4df20b02f", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.836625576019287, "target_z": -1.6796875436776237, "ce": 6.173524379730225, "genuineness": 4.118232727050781, "identity": 0.43618959188461304, "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/casual_not_formal/es/c96128ec8e8f366375103f17.mp3" }, { "key": "original-casual_not_formal-es-0-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "4b6028459677bf90f41f74fcc41ac4eadae9e160dbbf209355d5d695a6a3d5ab", "task_id": "casual_not_formal-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 1.9163693189620972, "target_z": -1.5078125521999293, "ce": 5.903473377227783, "genuineness": 4.344170093536377, "identity": 0.7510500550270081, "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/casual_not_formal/es/95abb651a8e43ea5ee26d459.mp3" }, { "key": "original-casual_not_formal-es-1-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "a082cfde846c1f89aa57b167723e1e0f2882e2496d4771fb4ac80bd5d7bdcb0e", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.6888861656188965, "target_z": -0.9277343214282393, "ce": 5.831989288330078, "genuineness": 4.570107460021973, "identity": 0.7390263080596924, "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/casual_not_formal/es/a7499985d6b0a89d5fcc77ac.mp3" }, { "key": "original-casual_not_formal-es-1-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "5c868ecdc7565759cb8f72ff7374f5c94eb5652a22f740954805a869e9941a0d", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.6539981365203857, "target_z": -1.066406177286667, "ce": 6.173524379730225, "genuineness": 4.570107460021973, "identity": 0.712787926197052, "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/casual_not_formal/es/5436914d81bf16ba8c74ea85.mp3" }, { "key": "original-casual_not_formal-es-1-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "70a24db9dabe6954f17085b64026a0ba1ff362d024ebdcd60de3e7c46bcdd349", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.9031970500946045, "target_z": -0.7246094047712486, "ce": 5.967014789581299, "genuineness": 4.68307638168335, "identity": 0.6927998065948486, "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/casual_not_formal/es/ee74f3927134489d700b1544.mp3" }, { "key": "original-casual_not_formal-es-1-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "24bcde1e534c5c789b8bc65bf3c898c9e431a2b918878cc41177a944697362a2", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.1025562286376953, "target_z": -0.572265623308825, "ce": 6.054384231567383, "genuineness": 4.68307638168335, "identity": 0.5971572399139404, "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/casual_not_formal/es/dc424c56eb0d0d2bb9515fa0.mp3" }, { "key": "original-casual_not_formal-es-1-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "7f9693ad3780c1df127ebdfe7ae7af22d59f0433095a5d9c26c81718e0d4a206", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.3250555992126465, "target_z": -1.2265624981607561, "ce": 6.09409761428833, "genuineness": 4.337893962860107, "identity": 0.5434749126434326, "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/casual_not_formal/es/765db98fcdc1df7a80488a74.mp3" }, { "key": "original-casual_not_formal-es-1-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "ba7c659c7ce25f48f253a74282b973d7ab2f442b057080dc4a86da15e964e116", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.197251796722412, "target_z": -0.5117187468745681, "ce": 5.871703147888184, "genuineness": 4.507347106933594, "identity": 0.5541702508926392, "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/casual_not_formal/es/b2fe791a0d320042ef6d004a.mp3" }, { "key": "original-casual_not_formal-es-1-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "9fb0680f8416723e1212d9b6d0b5dbf493d25b9417984009287c7f66041f6ffc", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.5941903591156006, "target_z": -0.9414062654326756, "ce": 5.855817794799805, "genuineness": 4.494794845581055, "identity": 0.5058490633964539, "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/casual_not_formal/es/8d157d967dbfd98f0c93b0ef.mp3" }, { "key": "original-casual_not_formal-es-1-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "3339448b6fe3443b2eff5fe2f4ae2d82b6915e05a707b2945bb689cb0f3c63bc", "task_id": "casual_not_formal-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "original_audio_reused": true, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 3.4489426612854004, "target_z": -0.39941408732258105, "ce": 6.149696350097656, "genuineness": 4.645420074462891, "identity": 0.6348564624786377, "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/casual_not_formal/es/cf317c08e5282672e31a036c.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "bc260bdcdc3638448599099418a8b6a8c3db69889981842d3e989ddab1d0688e", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.4347031116485596, "target_z": -1.0937500555070452, "ce": 6.006728172302246, "genuineness": 4.457139015197754, "identity": 0.7673436999320984, "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/casual_not_formal/fr/32425396c53b6b77881f9668.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "b4ba16f8b1375552ad4aa5477da0dc9e3a7745ac75d8e32f310ee9df27f068b6", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.763645648956299, "target_z": -0.8671875465157244, "ce": 6.117925643920898, "genuineness": 4.482243061065674, "identity": 0.778468132019043, "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/casual_not_formal/fr/1de576bb4a127710eadfa6e6.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "590c23960968ec5f78bef7a1680d7bf123fb7c91906d6a4ad0ea77fb0e1a9087", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.663966178894043, "target_z": -0.9121092944742843, "ce": 6.268836498260498, "genuineness": 4.595211505889893, "identity": 0.7199126482009888, "hypothesis": "La porte est encore ouverte et j'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "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/casual_not_formal/fr/0733462f9d5d16d0500a7b08.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "ce4c32e12295c4994930330da19cf55798e351f42eba0b1ebb3064a4b791c18a", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.3948311805725098, "target_z": -0.9472656797458979, "ce": 6.046441555023193, "genuineness": 4.507347106933594, "identity": 0.797222375869751, "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/casual_not_formal/fr/c30e8d8a3997ba6eef024a02.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "ed28624a4b3d1a382229e5ad109d87d87d5480c0c5450393cbc9c2d25a0f2120", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.813485622406006, "target_z": -0.898437482199838, "ce": 6.229123115539551, "genuineness": 4.582659721374512, "identity": 0.6856966018676758, "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/casual_not_formal/fr/df346d83a3d28df87a869e9b.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "e4222e1d263a68986a07a03d16956f07338e5f00f9ddcb799d7b68a4878eda3d", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.1057605743408203, "target_z": -1.417968730448636, "ce": 6.062326908111572, "genuineness": 4.118232727050781, "identity": 0.7502747774124146, "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/casual_not_formal/fr/be69b3ee08f82db5212f6105.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "5c2e1a3692ee8e40675ec074e24f12b54725aae9362f8d81faf183cbfee3da50", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.9430689811706543, "target_z": -0.7441405756389712, "ce": 6.157639026641846, "genuineness": 4.406930446624756, "identity": 0.6028651595115662, "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/casual_not_formal/fr/4ab4231383b0c21ca70858f4.mp3" }, { "key": "original-casual_not_formal-fr-0-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "6415c31305885f24453fbf10f611d49a9bf4a291b62bf6baa9a842b4d022a390", "task_id": "casual_not_formal-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.035984754562378, "target_z": -1.539062459568666, "ce": 6.165581703186035, "genuineness": 4.074300289154053, "identity": 0.5848194360733032, "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/casual_not_formal/fr/9ef963cba708321680a2b0db.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "b398b5f1b4e5599915db1f5cdbceba4a6d8445a64dc7501aa1708ed9b83181ff", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.9530367851257324, "target_z": -0.7539062585495209, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.6214693784713745, "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/casual_not_formal/fr/958eec508db60f0f195dfb63.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "8c2caee534c57a1c1f370b8a8657609fc55cc601f4cb67c296a33d2bbaa00bdf", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.569270610809326, "target_z": -1.126953002960053, "ce": 6.213237762451172, "genuineness": 4.507347106933594, "identity": 0.601410984992981, "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/casual_not_formal/fr/fc81ff44be66b9a295304e5d.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "4afad9f8446a1c447e92e6e072e5fb46592f3835980622af745d2f398f5695cf", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.7038378715515137, "target_z": -1.0429688338624454, "ce": 6.213237762451172, "genuineness": 4.319066047668457, "identity": 0.765318751335144, "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/casual_not_formal/fr/e0913f7d0cbf8025697b9c8a.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "5cbf68a471ea8d71db55823b12d1517250e0a39d25e1438ecb5f4513620a5d22", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.7102901935577393, "target_z": -0.2315673794039371, "ce": 6.0146708488464355, "genuineness": 4.720732688903809, "identity": 0.8336644172668457, "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/casual_not_formal/fr/c38616bfe424aa7ae5ab05f9.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "0126d28d093f0e4faa51c1860fc8eadb91887e82fbfb362c013ae6eb40233a14", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.0128445625305176, "target_z": -0.6562500931210897, "ce": 6.117925643920898, "genuineness": 4.545003414154053, "identity": 0.7608411312103271, "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/casual_not_formal/fr/c4a611bd86c3806d71eceda4.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "8d3b1ba8660ebcadca72ea3813be55caa178f854e97b25ea631200581d63c390", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.202235698699951, "target_z": -0.5859374803413901, "ce": 6.070269584655762, "genuineness": 4.808597087860107, "identity": 0.6271002292633057, "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/casual_not_formal/fr/732be0babf0deafef31f43ca.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "634173eea372bcad78923165ba3d58782435c9b85a9f332430571aaea60897d9", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.022812604904175, "target_z": -0.6093750164987632, "ce": 6.1020402908325195, "genuineness": 4.607763767242432, "identity": 0.7474722862243652, "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/casual_not_formal/fr/742893c22f168710e4b641bc.mp3" }, { "key": "original-casual_not_formal-fr-1-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "9c511729bce3075227113a9d50da1f2b3768fdc92c36a9aad1653ddf1ae47f86", "task_id": "casual_not_formal-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.242107629776001, "target_z": -0.6132812343352194, "ce": 6.252951145172119, "genuineness": 4.6203155517578125, "identity": 0.7030025124549866, "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/casual_not_formal/fr/82713efd6c22184bb08b1a39.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "1b09b34423d492e45a4cc7366c5133332f4d9283f963f3b1c46d1706b85b7aa4", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.2346317768096924, "target_z": -0.4453124680337447, "ce": 6.1020402908325195, "genuineness": 4.482243061065674, "identity": 0.5264029502868652, "hypothesis": "De de deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg, maar wanneer je klaar bent.", "wer": 0.05263157894736842, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/casual_not_formal/nl/efe8d4cc366142b737dfc726.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "ec832ee987f6eac7c1a12d3c2c1a13f67587e318eeb8fa1e45861f15c788ca4d", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.8932290077209473, "target_z": -0.49023444444863873, "ce": 6.0146708488464355, "genuineness": 4.306513786315918, "identity": 0.7503867149353027, "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/casual_not_formal/nl/6a75cb1ce4315164581218b3.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "93c42cce52ac8677dbc84af73de39121179798e6553ad2719a9ba09c99f98f0d", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.1449201107025146, "target_z": -0.3540039142558695, "ce": 6.054384231567383, "genuineness": 4.344170093536377, "identity": 0.5704206228256226, "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/casual_not_formal/nl/ef68ee1555a3321ef62c7dce.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "0aa4a15a42ada666fb2ece26bc94bf2c205cfa069ed1d19b0e9ca85923f3e6b1", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.813485622406006, "target_z": -0.6699218197930898, "ce": 6.125868320465088, "genuineness": 4.369274139404297, "identity": 0.7112761735916138, "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/casual_not_formal/nl/c94bce1da0edc9c775034aa0.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "2316de2dfb46d15c2f4d3f090c68aeffcf6e48e81955d4db2e83440ff5d1260c", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.823453426361084, "target_z": -0.6679687462382757, "ce": 6.133810997009277, "genuineness": 4.331617832183838, "identity": 0.6607411503791809, "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/casual_not_formal/nl/81cb67a66be7c0a9a899d95e.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "cce7fa6753f5f7e41fdc650439f9fc2960637926320c5c9bad1b5dad515809ce", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.295151710510254, "target_z": -1.0546875015943304, "ce": 6.1894097328186035, "genuineness": 4.582659721374512, "identity": 0.6109222173690796, "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/casual_not_formal/nl/474e9efe36c51bf5e93054a3.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "09d057371482115920cb2bddcb46698862a76bbc313db24679edbce58df395b8", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.3848633766174316, "target_z": -0.9824218473593278, "ce": 6.038498878479004, "genuineness": 4.193545341491699, "identity": 0.752799928188324, "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/casual_not_formal/nl/0aa91f4131266698a4d8c82c.mp3" }, { "key": "original-casual_not_formal-nl-0-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "603076200c7fca4dc8380f9c1f3485d4d725f54f14f743541e68534731bfb5ec", "task_id": "casual_not_formal-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.3118834495544434, "target_z": -0.3994140411610165, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.6901633143424988, "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/casual_not_formal/nl/f6981ab81ff3379386221b74.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1701-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "e9f7010a626dd11eafacc8280663807d0f2eadf1cf7ae74aa5bcb1a9fecae0b3", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.314375400543213, "target_z": -0.2939452731963817, "ce": 6.213237762451172, "genuineness": 4.344170093536377, "identity": 0.7185832858085632, "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/casual_not_formal/nl/37935abf22ffca3e838866d5.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1701-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "98bd7e66def8bc43b9ab3c5bbc7fd367e5f3cc6dd7fc6c6d14c77b86a9764617", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.344991445541382, "target_z": -1.0585937853281955, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.8254506587982178, "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/casual_not_formal/nl/4993d9353cd2426138350a07.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1701-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "c36a1dc6f2996f4efeff8869c6796ed581497c7201171f8ffe336cf6b7380fc1", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.247091770172119, "target_z": -0.5214843449654295, "ce": 6.1020402908325195, "genuineness": 4.425758361816406, "identity": 0.5977513194084167, "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/casual_not_formal/nl/10f5f8291168159fae3c62a3.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1701-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "5c2626116ea82b34e8962096efdb5f41bc0b5a12e88adf72d780782170c09c23", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.305119514465332, "target_z": -1.0898438727730837, "ce": 6.086154937744141, "genuineness": 4.362998008728027, "identity": 0.6767673492431641, "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/casual_not_formal/nl/e4840f3b946161c3f8575ef7.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1702-scale0.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1702, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "f53578bd35b8422ea2086c220ae4ac4a6e10b168d93d6349e284e8da727cfd56", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 3.6183979511260986, "target_z": -0.2949219191294315, "ce": 6.133810997009277, "genuineness": 4.545003414154053, "identity": 0.7498639822006226, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/casual_not_formal/nl/cd48934695050a84ec818829.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1702-scale0.5", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "7a25ab8dfbc598dd8883e07292dcc7014cf8473771b763fa280da81076d0cf3f", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 4.131747722625732, "target_z": 0.1105956317511784, "ce": 6.0305562019348145, "genuineness": 4.557555198669434, "identity": 0.7117646932601929, "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/casual_not_formal/nl/42bf34369a0c0d7185eec8a8.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1702-scale0.75", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "1714c4de204cb65101f562b6e0ed3dfccb40073c140c8598d558bc4c561386a2", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.9929087162017822, "target_z": -0.7109374737511671, "ce": 6.054384231567383, "genuineness": 4.457139015197754, "identity": 0.5766299962997437, "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/casual_not_formal/nl/c526b254f0c48c5ec3a6036e.mp3" }, { "key": "original-casual_not_formal-nl-1-seed1702-scale1.0", "bucket": "casual_not_formal", "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. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": [ "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Very informal/casual, not formal. 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": "c0a4c1f8a354130caa6c315c42097e02a64ce8fa050d366d19ef3715a74baf9c", "mp3_sha256": "43dfd2e4d03683fa48d817f3a981d37bc49d6ea9b2f2fe4303f6e2a3ca81e73e", "task_id": "casual_not_formal-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Very informal/casual, not formal. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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(Very informal/casual, not formal. 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": 2.3150875568389893, "target_z": -1.1953125710885892, "ce": 6.221180438995361, "genuineness": 4.337893962860107, "identity": 0.7326463460922241, "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/casual_not_formal/nl/a7a7695dd2021406a33821e0.mp3" } ], "source_standard_audio_unchanged": true }