{ "complete": true, "report": { "bucket": "emo_Pain", "label": "Intense Pain", "definition": { "id": "emo_Pain", "category": "emotion", "head": "emo_Pain", "pole": "high", "label": "Intense Pain", "floor": 3.0 }, "recipe_status": "direct_manual_recipe", "manual_provenance": [ { "id": "rank4/emotion_pain/full-v2-sealed/sealed-selected", "mission": "emotion_pain", "format_status": "CANONICAL", "heldout": true, "qualified": true, "source_cfg": 3, "source_model": { "label": "Humaneness 1.5 + real-speech rank 4", "repo": "laion/Humaneness-Voice-Large-1.5-RealSpeech-LoRAs", "revision": "fed33e554f094fcdc89fc0124aed2f9d42018714", "rank": 4 }, "metrics": { "ce": 6.205136222839355, "genuineness": 4.480360164642334, "wer": 0.016077008055178244, "identity": 0.6593440943956375, "voiceclap": 0.07878162510693074, "fitness": 0.622770754836809, "steering_v2": 0.5842315468243826, "raw_target": 0.44260206282138825, "target_cdf": 0.5842315468243826, "family_hit_all": null, "nominal_hit_all": null, "n": 100, "valid_fraction": 1, "by_arm": { "reference": { "n": 50, "ce": 6.214508581161499, "genuineness": 4.4600258064270015, "wer": 0.016955424582232145, "identity": 0.6593440943956375, "fitness": 0.6218949800217468, "steering_v2": 0.5849250094617239 }, "none": { "n": 50, "ce": 6.195763864517212, "genuineness": 4.500694522857666, "wer": 0.015198591528124342, "identity": null, "fitness": 0.6236465296518713, "steering_v2": 0.5835380841870413 } } } } ], "selected_wording": { "id": "manual-abbf6e2776b7", "general": "The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.", "inline": [ "Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.", "Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.", "Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress." ], "cfg": 3, "recipe": "P06", "strategy": "local-Gemma-mutation", "timing_syntax": "canonical", "canonical_labels": true, "neutral_event": "keep", "parent": "cfg-3", "origin": "exact_frozen_manual_recipe" }, "selection": { "manual_no_adapter": { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 0.39243163288316063, "target_raw": 0.48268012031912805, "ce": 5.944378280639649, "genuineness": 4.575128293037414, "identity": 0.559913519024849, "wer": 0.005572755417956656, "valid_fraction": 1, "identity_delta": -0.007984544336795806, "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": 0.9853027279487674, "target_raw": 0.8236827306449414, "ce": 5.736379623413086, "genuineness": 4.451490378379821, "identity": 0.5308944940567016, "wer": 0.014298245614035088, "valid_fraction": 1, "identity_delta": -0.037003569304943085, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, "grid": [ { "config": { "scale": 0.0, "cfg": 3.0 }, "n": 20, "target_z": 0.39243163288316063, "target_raw": 0.48268012031912805, "ce": 5.944378280639649, "genuineness": 4.575128293037414, "identity": 0.559913519024849, "wer": 0.005572755417956656, "valid_fraction": 1, "identity_delta": -0.007984544336795806, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.5 }, "n": 20, "target_z": 0.12203368694077729, "target_raw": 0.32715490385890006, "ce": 6.028769183158874, "genuineness": 4.602429151535034, "identity": 0.5844735354185104, "wer": 0.013742690058479532, "valid_fraction": 1, "identity_delta": 0.016575472056865694, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 2.0 }, "n": 20, "target_z": -0.13785400943938814, "target_raw": 0.17767488434910775, "ce": 6.0885377645492555, "genuineness": 4.613098359107971, "identity": 0.6362082853913307, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.06831022202968598, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.5 }, "n": 20, "target_z": -0.2671386701849514, "target_raw": 0.1033140204846859, "ce": 6.027379155158997, "genuineness": 4.556300210952759, "identity": 0.6110758021473884, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.04317773878574371, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.25 }, "n": 20, "target_z": -0.2724609415885008, "target_raw": 0.10025280117988586, "ce": 6.093700504302978, "genuineness": 4.580776786804199, "identity": 0.5881297275424003, "wer": 0.010818713450292398, "valid_fraction": 1, "identity_delta": 0.020231664180755615, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 1.0 }, "n": 20, "target_z": -0.3271972642790332, "target_raw": 0.06877002343535424, "ce": 6.163993167877197, "genuineness": 4.560379695892334, "identity": 0.6203296542167663, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.05243159085512161, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 5.0 }, "n": 20, "target_z": 1.3284423797107041, "target_raw": 1.0210469096899033, "ce": 5.789396953582764, "genuineness": 4.384336709976196, "identity": 0.4910130716860294, "wer": 0.017068713450292396, "valid_fraction": 1, "identity_delta": -0.07688499167561531, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.0, "cfg": 4.0 }, "n": 20, "target_z": 1.3207519540864048, "target_raw": 1.0166235953569411, "ce": 5.841123580932617, "genuineness": 4.401909637451172, "identity": 0.50718799456954, "wer": 0.005756578947368421, "valid_fraction": 1, "identity_delta": -0.06071006879210472, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 4.0 }, "n": 20, "target_z": 0.9853027279487674, "target_raw": 0.8236827306449414, "ce": 5.736379623413086, "genuineness": 4.451490378379821, "identity": 0.5308944940567016, "wer": 0.014298245614035088, "valid_fraction": 1, "identity_delta": -0.037003569304943085, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 3.0 }, "n": 20, "target_z": 0.6410400244440857, "target_raw": 0.6256726041436196, "ce": 5.7841349124908445, "genuineness": 4.484439539909363, "identity": 0.5655783817172051, "wer": 0.0029411764705882353, "valid_fraction": 1, "identity_delta": -0.0023196816444396974, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 3.0 }, "n": 20, "target_z": 0.47152099120829516, "target_raw": 0.528170072287321, "ce": 5.839932131767273, "genuineness": 4.484439563751221, "identity": 0.528684151917696, "wer": 0.019269005847953217, "valid_fraction": 0.95, "identity_delta": -0.03921391144394874, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 3.0 }, "n": 20, "target_z": 0.41030272535803414, "target_raw": 0.4929590649902821, "ce": 5.890368175506592, "genuineness": 4.565086674690247, "identity": 0.5916319563984871, "wer": 0.01715686274509804, "valid_fraction": 0.9, "identity_delta": 0.023733893036842348, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.5 }, "n": 20, "target_z": 0.3896972629836666, "target_raw": 0.4811073884367943, "ce": 5.888382482528686, "genuineness": 4.533078861236572, "identity": 0.5310828655958175, "wer": 0.003125, "valid_fraction": 1, "identity_delta": -0.03681519776582718, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.5 }, "n": 20, "target_z": 0.27019042173819957, "target_raw": 0.41237044855952265, "ce": 5.885404014587403, "genuineness": 4.471573734283448, "identity": 0.5463211894035339, "wer": 0.0254265565875473, "valid_fraction": 1, "identity_delta": -0.02157687395811081, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 2.0 }, "n": 20, "target_z": 0.23104247320508306, "target_raw": 0.38985366076231004, "ce": 5.971581935882568, "genuineness": 4.565714311599732, "identity": 0.5570998921990394, "wer": 0.005555555555555555, "valid_fraction": 1, "identity_delta": -0.010798171162605286, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 2.0 }, "n": 20, "target_z": 0.11593017217598167, "target_raw": 0.3236443355679512, "ce": 5.968404912948609, "genuineness": 4.557869243621826, "identity": 0.5784263789653779, "wer": 0.013797514619883041, "valid_fraction": 0.95, "identity_delta": 0.010528315603733063, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.5 }, "n": 20, "target_z": 0.05954589447549984, "target_raw": 0.2912137009203434, "ce": 5.986474466323853, "genuineness": 4.58171820640564, "identity": 0.5738860458135605, "wer": 0.022624269005847953, "valid_fraction": 0.95, "identity_delta": 0.005987982451915741, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.5 }, "n": 20, "target_z": -0.05130615440065047, "target_raw": 0.2274547517299652, "ce": 6.031946229934692, "genuineness": 4.586739087104798, "identity": 0.5520555336028338, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.015842529758810998, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 2.0 }, "n": 20, "target_z": -0.057702643521954596, "target_raw": 0.22377567291259765, "ce": 5.954902315139771, "genuineness": 4.56163489818573, "identity": 0.587873800098896, "wer": 0.010818713450292398, "valid_fraction": 1, "identity_delta": 0.019975736737251282, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.5 }, "n": 20, "target_z": -0.13779297435650853, "target_raw": 0.17770998999476434, "ce": 6.071659636497498, "genuineness": 4.5208406209945675, "identity": 0.5852382943034172, "wer": 0.00849673202614379, "valid_fraction": 1, "identity_delta": 0.01734023094177246, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.5 }, "n": 20, "target_z": -0.17717284791422144, "target_raw": 0.1550598055124283, "ce": 6.055575704574585, "genuineness": 4.566969466209412, "identity": 0.6318533934652806, "wer": 0.008742690058479531, "valid_fraction": 1, "identity_delta": 0.06395533010363579, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.25 }, "n": 20, "target_z": -0.2200195292057496, "target_raw": 0.1304156117141247, "ce": 6.069078207015991, "genuineness": 4.5842286348342896, "identity": 0.5607663869857789, "wer": 0.006666666666666666, "valid_fraction": 1, "identity_delta": -0.0071316763758659365, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.25 }, "n": 20, "target_z": -0.2397827157267245, "target_raw": 0.1190483883023262, "ce": 6.079006552696228, "genuineness": 4.551593232154846, "identity": 0.5629926092922688, "wer": 0.010982112143102855, "valid_fraction": 1, "identity_delta": -0.004905454069375992, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.25 }, "n": 20, "target_z": -0.25189209534146795, "target_raw": 0.11208341717720031, "ce": 6.1500934839248655, "genuineness": 4.5387273073196415, "identity": 0.5881036818027496, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.02020561844110489, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 1.0 }, "n": 20, "target_z": -0.3076904295103991, "target_raw": 0.07998980060219765, "ce": 6.136193776130677, "genuineness": 4.505464315414429, "identity": 0.6017800301313401, "wer": 0.005718954248366013, "valid_fraction": 1, "identity_delta": 0.03388196676969528, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 1.0 }, "n": 20, "target_z": -0.31391601573632055, "target_raw": 0.07640902027487755, "ce": 6.135002398490906, "genuineness": 4.507974743843079, "identity": 0.5484503000974655, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": -0.01944776326417923, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 1.0 }, "n": 20, "target_z": -0.34443359603713164, "target_raw": 0.05885617509484291, "ce": 6.158036136627198, "genuineness": 4.5123679637908936, "identity": 0.5843412384390831, "wer": 0.002631578947368421, "valid_fraction": 1, "identity_delta": 0.016443175077438355, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 5.0 }, "n": 20, "target_z": 1.670556607623512, "target_raw": 1.21782129406929, "ce": 5.709970164299011, "genuineness": 4.3473081350326535, "identity": 0.47654592506587506, "wer": 0.03920687134502924, "valid_fraction": 0.9, "identity_delta": -0.09135213829576969, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.75, "cfg": 5.0 }, "n": 20, "target_z": 1.535156205147932, "target_raw": 1.139942829310894, "ce": 5.624387907981872, "genuineness": 4.282351124286651, "identity": 0.5120633022859693, "wer": 0.03777412280701754, "valid_fraction": 0.85, "identity_delta": -0.05583476107567549, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 5.0 }, "n": 20, "target_z": 1.5271484069374446, "target_raw": 1.135336971282959, "ce": 5.59450364112854, "genuineness": 4.332559370994568, "identity": 0.4790778923779726, "wer": 0.014583333333333334, "valid_fraction": 0.95, "identity_delta": -0.08882017098367215, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 0.5, "cfg": 4.0 }, "n": 20, "target_z": 1.415673813946808, "target_raw": 1.071219952404499, "ce": 5.755392384529114, "genuineness": 4.326283276081085, "identity": 0.5127997286617756, "wer": 0.025603070175438596, "valid_fraction": 0.95, "identity_delta": -0.055098334699869154, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" }, { "config": { "scale": 1.0, "cfg": 4.0 }, "n": 20, "target_z": 1.315429674910656, "target_raw": 1.0135623715817927, "ce": 5.666980481147766, "genuineness": 4.359546375274658, "identity": 0.5012681705877184, "wer": 0.008680555555555556, "valid_fraction": 1, "identity_delta": -0.06662989277392625, "identity_pairs": 20, "gates_pass": false, "gate_scope": "development_provisional_20_pairs" } ], "development_standard": [ { "scale": 0.0, "n": 20, "target_z": -0.3352539068774432, "target_raw": 0.06413607150316239, "ce": 6.16955304145813, "genuineness": 4.455883574485779, "identity": 0.5678980633616447, "wer": 0.0058823529411764705, "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": -0.33251952791038303, "target_raw": 0.06570880860090256, "ce": 6.148703503608703, "genuineness": 4.4106961488723755, "identity": 0.6181595355272294, "wer": 0.0, "valid_fraction": 1, "identity_delta": 0.05026147216558456, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 0.75, "n": 20, "target_z": -0.3229980497996949, "target_raw": 0.07118529230356216, "ce": 6.17868709564209, "genuineness": 4.45400083065033, "identity": 0.5852677941322326, "wer": 0.007894736842105263, "valid_fraction": 1, "identity_delta": 0.01736973077058792, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" }, { "scale": 1.0, "n": 20, "target_z": -0.3395507803391954, "target_raw": 0.06166463196277618, "ce": 6.152873420715332, "genuineness": 4.353584122657776, "identity": 0.5671944677829742, "wer": 0.0, "valid_fraction": 1, "identity_delta": -0.0007035955786705017, "identity_pairs": 20, "gates_pass": true, "gate_scope": "development_provisional_20_pairs" } ] }, "heldout": { "standard": { "n": 50, "target_z": -0.34390625048576245, "target_raw": 0.059159489274024965, "ce": 6.221180419921875, "genuineness": 4.423499188423157, "identity": 0.5939027291536331, "wer": 0.010232072275550536, "valid_fraction": 1, "identity_delta": 0.0, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "manual": { "n": 50, "target_z": 0.14971679607551594, "target_raw": 0.3430774417519569, "ce": 6.0271409034729, "genuineness": 4.489146618843079, "identity": 0.6024945497512817, "wer": 0.015677608051749926, "valid_fraction": 1, "identity_delta": 0.00859182059764862, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "adapter": { "n": 50, "target_z": 0.8902636697963414, "target_raw": 0.7690189644694329, "ce": 5.798789148330688, "genuineness": 4.239485726356507, "identity": 0.5498610246554017, "wer": 0.047060982495765104, "valid_fraction": 0.98, "identity_delta": -0.04404170449823141, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" }, "matched_no_adapter": { "n": 50, "target_z": 0.5274609338163012, "target_raw": 0.5603451380133628, "ce": 5.921821041107178, "genuineness": 4.387725777626038, "identity": 0.5914051773771644, "wer": 0.016404103143233578, "valid_fraction": 1, "identity_delta": -0.0024975517764687536, "identity_pairs": 50, "gates_pass": true, "gate_scope": "heldout_50_reference_pairs" } }, "adapter_vs_matched_cfg": { "n": 50, "delta_z": 0.3628027359800402, "ci95_z": [ 0.21553710562046097, 0.5911914129008978 ], "language_deltas": { "en": 0.3277343841485265, "de": 0.2131347658982414, "fr": 0.28481445376256326, "nl": 0.8050293048192404, "es": 0.1833007712716293 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.12303189277648925, "delta_genuineness": -0.14824005126953124, "delta_wer": 0.030656879352531526, "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": 0.7405468737208255, "ci95_z": [ 0.5923535132103558, 0.9177441543214776 ], "language_deltas": { "en": 0.6820800981121017, "de": 0.504492178472914, "fr": 0.6509277363686502, "nl": 1.0748535251277282, "es": 0.7903808305227332 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.22835175514221193, "delta_genuineness": -0.24966089248657228, "delta_wer": 0.03138337444401518, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "manual_vs_standard": { "n": 50, "delta_z": 0.49362304656127837, "ci95_z": [ 0.2824414079785048, 0.704804685144052 ], "language_deltas": { "en": 0.2147949180799807, "de": 0.26743164972292627, "fr": 0.7051757693285794, "nl": 0.44775390428670997, "es": 0.8329589913881956 }, "languages_improved": 5, "clear_gain": true, "delta_ce": -0.1940395164489746, "delta_genuineness": 0.06564743041992188, "delta_wer": 0.005445535776199391, "note": "Exploratory language-cluster bootstrap; five voices, one fixed reference voice per language. No familywise claim across72targets." }, "route": { "adapter": true, "scale": 0.75, "cfg": 4.0, "reason": "Held-out target improves against matched CFG and best manual-only configuration; QA passes" }, "reference_only": true, "updated": 1793024911.624254, "example": { "text": "Your message arrived after I had left. I read it twice... before putting the phone away. Let me know when we can talk.", "sentences": [ "Your message arrived after I had left.", "I read it twice... before putting the phone away.", "Let me know when we can talk." ], "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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.", "language": "en", "seed": 1702, "scale": 0.75, "cfg": 4.0, "frames": 128, "reference": true } }, "rows": [ { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "0026da7bdc35d1ac6ecff002c28b3b442c3c7b5a7b2517c385f1e1704498201f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e33274e4d0d1b42e0d2a81ec2f970c196f2e2b89be379965c17105e4d91bbb08", "generation_seconds_batch": 23.27977334696334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9040325880050659, "target_z": 1.1249999222780047, "ce": 5.772419452667236, "genuineness": 4.457139015197754, "identity": 0.48499345779418945, "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/emo_Pain/0026da7bdc35d1ac6ecff002c28b3b442c3c7b5a7b2517c385f1e1704498201f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "00463b5c5631ea829cbc0b42d2e12d08c905b0b991d929aa56f2076ec3f6b138", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0cbf149787779d4c8484f4dd7e1cc9bf01617cf60bfbf5d6c5c4a94808cc94fa", "generation_seconds_batch": 16.934556304011494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 6.340320587158203, "genuineness": 4.582659721374512, "identity": 0.6378741264343262, "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/emo_Pain/00463b5c5631ea829cbc0b42d2e12d08c905b0b991d929aa56f2076ec3f6b138.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "004ad1f63a0f10cd54e2d553712999c9fa3e2cb13eb71a5848bebf3917cc08ab", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f672a35ce351f081408167e93bb64c5398e153a57d66abe220bd7152d4a866ec", "generation_seconds_batch": 22.30763688602019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15923035144805908, "target_z": -0.16992188835846794, "ce": 5.760505676269531, "genuineness": 4.055472373962402, "identity": 0.38182777166366577, "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/emo_Pain/004ad1f63a0f10cd54e2d553712999c9fa3e2cb13eb71a5848bebf3917cc08ab.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "006f57bcde93929d25b876ce70eaffa6da64330437c1afb49619fe695af45e20", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4fb92ec840c16824a699d79d9d8be5545783e6d70f92de3fc9c2be997cd9beb5", "generation_seconds_batch": 23.038734287023544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 5.95112943649292, "genuineness": 4.7081804275512695, "identity": 0.578772246837616, "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/emo_Pain/006f57bcde93929d25b876ce70eaffa6da64330437c1afb49619fe695af45e20.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "00fc53d2cd5bd04b5d62ffb6c1d5af950a0100fb73b1411a57f9e9f691d71d80", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2228035c91bf2a96c5eafad00adab98bcf77ab0b7ff98c6b284ab85a3e99b6e7", "generation_seconds_batch": 16.74245912209153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09744435548782349, "target_z": -0.27734373785593824, "ce": 6.284721851348877, "genuineness": 4.519899368286133, "identity": 0.5121924877166748, "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/emo_Pain/00fc53d2cd5bd04b5d62ffb6c1d5af950a0100fb73b1411a57f9e9f691d71d80.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": "0178b68a2dd4fce453b8588ed53f324be2022bda5dc0028e5f5272e6d8ce1fc1", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "72435df42500c2b34977b5e53a2996c1ab031d5491ffa111c9b413916be770fc", "generation_seconds_batch": 21.03086467203684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1781378984451294, "target_z": 1.601562595533286, "ce": 6.157639026641846, "genuineness": 3.9801597595214844, "identity": 0.655134916305542, "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/emo_Pain/0178b68a2dd4fce453b8588ed53f324be2022bda5dc0028e5f5272e6d8ce1fc1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "01a23de4a4d7cb8484ee4d4680dcb766d0942805607624414cfa090d5ddd84d0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e5afc5184bd83c7b4885c84ef6486bcab9e6e640d6d0868b53feb7477068a888", "generation_seconds_batch": 22.08014843193814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22888004779815674, "target_z": -0.048828118118365, "ce": 5.637393951416016, "genuineness": 4.6705241203308105, "identity": 0.6847656965255737, "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/emo_Pain/01a23de4a4d7cb8484ee4d4680dcb766d0942805607624414cfa090d5ddd84d0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "01e5186a03a69b83f8d4bfe61aeed34da0d80fba7986ca2fa6c6d414025fda7e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b0eedb6445157249010bfd8178aa81532f87ceabb09a2da7c22f45f4b69f4459", "generation_seconds_batch": 23.74732047901489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6793562173843384, "target_z": 0.7343749773310847, "ce": 5.80021858215332, "genuineness": 4.770940780639648, "identity": 0.7733834981918335, "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/emo_Pain/01e5186a03a69b83f8d4bfe61aeed34da0d80fba7986ca2fa6c6d414025fda7e.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "e7fb0bac41ac56a311861d0b7c26f223762ea537bb2e79fe6ad0adb72eb6b9a1", "generation_seconds_batch": 16.467003004974686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.032288193702697754, "target_z": -0.3906249967615835, "ce": 5.919358730316162, "genuineness": 4.595211505889893, "identity": 0.7153352499008179, "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/emo_Pain/01f5d5f22d9b51f4e674eec92be80d0710e62f5704093e3409bc811236eec3e5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "02fc7862df50e571f50399fb6ae789d9e980054bd1cb5ab1eb3f44ba9be95472", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "da1ea8f54f83644925f7ea3309cf9770d8ac6792daa2f5efb1912db9ac513651", "generation_seconds_batch": 22.942274622037075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8725779056549072, "target_z": 1.0703124465661282, "ce": 5.6254801750183105, "genuineness": 4.394378662109375, "identity": 0.5720552206039429, "hypothesis": "J'ai tout, j'ai trouvé 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/emo_Pain/02fc7862df50e571f50399fb6ae789d9e980054bd1cb5ab1eb3f44ba9be95472.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "0391a013435878af569f628f2418ed0e0cfb0931699ae1c23c5e7e335be10084", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "101423cf4842ae07e4711fd144267217f811e6c34ac53804359ab0961ab20054", "generation_seconds_batch": 22.369334277929738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9444743394851685, "target_z": 1.1953124206587964, "ce": 5.8875885009765625, "genuineness": 4.65797233581543, "identity": 0.5981469750404358, "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/emo_Pain/0391a013435878af569f628f2418ed0e0cfb0931699ae1c23c5e7e335be10084.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "039936553209a9f0e8f9fa1e4f10a2a597175468011afffa32680abda1bdf1d4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ac0668824dd753416e9ddb4008aa0e8ee4a6e23ea83729303c06d4520ad75000", "generation_seconds_batch": 21.487092609051615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.09409761428833, "genuineness": 4.344170093536377, "identity": 0.5618282556533813, "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/emo_Pain/039936553209a9f0e8f9fa1e4f10a2a597175468011afffa32680abda1bdf1d4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "0399756f7e9894b5c923aee4a010f1b5756b5ec828c9b85a7b090bd5cf0b61ff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b3c7f3c70cc45ddcc33665254d23549dc9db27b6455a3b0ca7d1a392e928d641", "generation_seconds_batch": 24.066187088028528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08396376669406891, "target_z": -0.3007812459519794, "ce": 6.022613525390625, "genuineness": 4.7081804275512695, "identity": 0.8093557357788086, "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/emo_Pain/0399756f7e9894b5c923aee4a010f1b5756b5ec828c9b85a7b090bd5cf0b61ff.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "03af04748322f308cf114c0bfb6455cc9bd7eff116066ab7bffbce5a22598e13", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "baca8ccfe61aac8b7bde55dcbf762d7be91ae52a703427535b8f12141048f8a6", "generation_seconds_batch": 22.27053001697641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1601637601852417, "target_z": 1.5703125501954553, "ce": 5.530168056488037, "genuineness": 4.369274139404297, "identity": 0.3736477196216583, "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": 0.120000000000001, "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/emo_Pain/03af04748322f308cf114c0bfb6455cc9bd7eff116066ab7bffbce5a22598e13.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "03cbd92b4d1e291c3304ebe629aa34095a21b723008b8697b669265ef92ee6e0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1701, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ca1f904ccdf7170414220c13680754c575eaea872c0b0e65385084d76fb08e2f", "generation_seconds_batch": 21.989109852002002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3758530616760254, "target_z": 1.94531247247346, "ce": 5.736677646636963, "genuineness": 4.306513786315918, "identity": 0.7491664886474609, "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/emo_Pain/03cbd92b4d1e291c3304ebe629aa34095a21b723008b8697b669265ef92ee6e0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "040c63704fa0c9a1e54627f874a243440efa84d3ea76821f5d912fb7658f39ba", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6d7d9b8a977b7ebe5e6ca20a5968df15be8a22eda3483001b6678c1590f4f240", "generation_seconds_batch": 21.857697119936347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21764622628688812, "target_z": -0.06835937054717736, "ce": 5.8875885009765625, "genuineness": 4.406930446624756, "identity": 0.14229008555412292, "hypothesis": "Er staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/040c63704fa0c9a1e54627f874a243440efa84d3ea76821f5d912fb7658f39ba.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "04bc569a0a9692e5a8a8a2c3e66ad7b1c91e846e5a23c2fb9e55cd79d303671d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2d3b0fd54a3dc1a43f66fbc959c7574fc789c90e385c41bbeae4af8025f5b4cb", "generation_seconds_batch": 23.555264313006774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3803465366363525, "target_z": 1.9531248801785905, "ce": 5.2918877601623535, "genuineness": 3.8922953605651855, "identity": 0.02567952498793602, "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/emo_Pain/04bc569a0a9692e5a8a8a2c3e66ad7b1c91e846e5a23c2fb9e55cd79d303671d.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": "04d792cf0464d43401465310c0ccec6ccd17b61e0200bba30ff51f8f049c6649", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f83d0b0b1fef82bd067ed17bbf0ac2db35e56d8275fca7b201d8cd51bf1270d2", "generation_seconds_batch": 18.4809860419482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3614391088485718, "target_z": 0.1816406035454909, "ce": 6.078212261199951, "genuineness": 4.344170093536377, "identity": 0.6829483509063721, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/04d792cf0464d43401465310c0ccec6ccd17b61e0200bba30ff51f8f049c6649.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "057dbcb5d090108b4904be793c539716c27af70e5c056d2ae9af6e50593bd7d2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8ce58766575dc6b6fbb2624056c8858b8bd47a710881d88b143189c17123c2ab", "generation_seconds_batch": 24.02667941409163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18619152903556824, "target_z": -0.12304687216638559, "ce": 5.855817794799805, "genuineness": 4.507347106933594, "identity": 0.6383544206619263, "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/emo_Pain/057dbcb5d090108b4904be793c539716c27af70e5c056d2ae9af6e50593bd7d2.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "3b3c1608d464bd834d5e32e1b999092661796a523986955347c72a4199513980", "generation_seconds_batch": 17.93150353396777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.284721851348877, "genuineness": 4.632867813110352, "identity": 0.6549292802810669, "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/emo_Pain/059867605942d1f4c46c2c44d936e08b7fbe89dd4feef4c05500a7ed426c27a3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "05bdffa163ea5e0f3fcdb6488aac3b4ae5f309f76b0786bf41675446e3164f8c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "37c1a5e1cfa2cc7b3d1d6d2721d003e8e01e189e43e0b5dcb760bd0dfa44f420", "generation_seconds_batch": 24.43903627502732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09744435548782349, "target_z": -0.27734373785593824, "ce": 6.141753673553467, "genuineness": 4.645420074462891, "identity": 0.692004919052124, "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/emo_Pain/05bdffa163ea5e0f3fcdb6488aac3b4ae5f309f76b0786bf41675446e3164f8c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "0691924834581667b839159004adf1af4951374b47ddaf540ee0db4b51b4560e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "74538c5d057cda93ae65acdb425fc320b4ed045af3d672b79db2ac2df3dc83e3", "generation_seconds_batch": 21.723850366077386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13563933968544006, "target_z": -0.2109374951423753, "ce": 6.252951145172119, "genuineness": 4.8588056564331055, "identity": 0.7882812023162842, "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/emo_Pain/0691924834581667b839159004adf1af4951374b47ddaf540ee0db4b51b4560e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "06bffada860bae1eb13b4c8e11c544d50cd1dbc2491b5db877e4712d5a9b71a2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "96ff847bb6581ff17642287b439d9e6bbe9987ddea100ab86d93227f2f86b033", "generation_seconds_batch": 22.96319531800691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5960359573364258, "target_z": 2.3281250097152495, "ce": 5.383228778839111, "genuineness": 4.356721878051758, "identity": 0.6012648940086365, "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/emo_Pain/06bffada860bae1eb13b4c8e11c544d50cd1dbc2491b5db877e4712d5a9b71a2.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": "0763724d559dc436e40a9b7c615567eaf826be5160670ea8ae66f237939bf66b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b32541178420e79859addc22b12296b69f819f6b97dee540749d99e4a1756cbf", "generation_seconds_batch": 17.465707438997924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1457497775554657, "target_z": -0.19335937054717736, "ce": 6.054384231567383, "genuineness": 4.7834930419921875, "identity": 0.5948079824447632, "hypothesis": "Mes 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": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0763724d559dc436e40a9b7c615567eaf826be5160670ea8ae66f237939bf66b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "076a93534c66b815c61e373a7fb0103a2335a70e3cf6e35eeb1bb2e57dc15004", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "552d9fa854d73fb75d9236e859d2e8faddecc9308824883555099d9dc600dc4d", "generation_seconds_batch": 21.51887546305079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4221017360687256, "target_z": 0.2871093511166785, "ce": 5.673136234283447, "genuineness": 4.381826400756836, "identity": 0.46394598484039307, "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/emo_Pain/076a93534c66b815c61e373a7fb0103a2335a70e3cf6e35eeb1bb2e57dc15004.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "0877d2a722b1a651b4c8b952e63975e538f6e6cb898fae7d7d17f6b5755b9132", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f2eec72cc46f580ff18a59b900e1a9c6731587c631b21f0ae87e10031840e31f", "generation_seconds_batch": 21.84408919804264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2311268150806427, "target_z": -0.044921862451136184, "ce": 5.824047088623047, "genuineness": 4.7458367347717285, "identity": 0.683638870716095, "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/emo_Pain/0877d2a722b1a651b4c8b952e63975e538f6e6cb898fae7d7d17f6b5755b9132.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "09ad38d817e1bbb84a88cb896c490a950ae874375c198331c0319acbc85cf1da", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f233f2a4f5639c3280a3ad020f4eb7ac71218dd265ca5e639c7ecb878e41d8bd", "generation_seconds_batch": 23.52213394199498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2069740891456604, "target_z": -0.08691407201284838, "ce": 5.824047088623047, "genuineness": 4.7332844734191895, "identity": 0.7305737733840942, "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/emo_Pain/09ad38d817e1bbb84a88cb896c490a950ae874375c198331c0319acbc85cf1da.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "09f3d740176235e340c391c8267de7692fb9b8510741d2a03d918c55cc9f272e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ef8fc6f9ac642c4f5bfbc6ffe0f7ea7f60976188f4ead438151f70636fa4c58d", "generation_seconds_batch": 24.521255424944684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05138568580150604, "target_z": -0.3574218754048021, "ce": 6.09409761428833, "genuineness": 4.457139015197754, "identity": 0.5989065170288086, "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/emo_Pain/09f3d740176235e340c391c8267de7692fb9b8510741d2a03d918c55cc9f272e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "0a439d92f30d07a2b9f639e401ffaecd2bd560ec7dcc83519278d28c63830e29", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4949b729d460a8ca16c40e3676f17ddd936d3abd79e2396bc537f0cbd6788d40", "generation_seconds_batch": 24.110118541982956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.046441555023193, "genuineness": 4.607763767242432, "identity": 0.5696038007736206, "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/emo_Pain/0a439d92f30d07a2b9f639e401ffaecd2bd560ec7dcc83519278d28c63830e29.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "0a5b36b9fa9b6c0fa2466a6845181723424784a9a3c24ff3a464528826483f79", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2fcb32f69655695cdb9e8652965c19ab0f1dd205b132dcd2e42cb2631092436a", "generation_seconds_batch": 18.148460026015528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19124674797058105, "target_z": -0.11425780986878661, "ce": 5.967014789581299, "genuineness": 4.7834930419921875, "identity": 0.7406611442565918, "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/emo_Pain/0a5b36b9fa9b6c0fa2466a6845181723424784a9a3c24ff3a464528826483f79.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "0a63e9efbf37b4bd6715455c306c600e76a9cb0b3a5175b994b92134ebd3e2e2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b330a97208bc2c00ed7cc84103db95601d90c87a4b7934b62e0ee9c5167739d6", "generation_seconds_batch": 23.917818334070034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31875061988830566, "target_z": 0.10742190131213382, "ce": 6.038498878479004, "genuineness": 4.124508857727051, "identity": 0.469967782497406, "hypothesis": "Ik heb het ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er samen in kijken.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0a63e9efbf37b4bd6715455c306c600e76a9cb0b3a5175b994b92134ebd3e2e2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "0ab1caedf351152f29755b1528a2872fce6a60862ebd3808d1ae350f0634a4fe", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a83d03ccaaa5e6558137e70ec2f69cfa27fe42291a2c08faf180a4ab42c50866", "generation_seconds_batch": 17.494559879996814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02891804277896881, "target_z": -0.39648438026242677, "ce": 5.967014789581299, "genuineness": 4.444586753845215, "identity": 0.5018208026885986, "hypothesis": "De deur staat nog open. Ik heb ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0ab1caedf351152f29755b1528a2872fce6a60862ebd3808d1ae350f0634a4fe.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "0ac030f41d3097beac0b196252a50dc70c977044fe99db2c0e7dc42e8f5839a6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "50357839fee4678003651c0c6907008291a17756c58d1f53b01f6de5604b53a4", "generation_seconds_batch": 15.7954231480835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 6.23706579208374, "genuineness": 4.444586753845215, "identity": 0.4701005816459656, "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/emo_Pain/0ac030f41d3097beac0b196252a50dc70c977044fe99db2c0e7dc42e8f5839a6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "0ad5a8ed3540d765e3f869b6a7d31b2cde92067450cedf2e08a8d6ac085360ef", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9e186d1532dcb7b6a60cbb3efd8a385764bda674d3e7dc03647d8e1565dac84c", "generation_seconds_batch": 23.98794989695307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2949695587158203, "target_z": 1.8046874757118765, "ce": 5.3792572021484375, "genuineness": 4.519899368286133, "identity": 0.5589838027954102, "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/emo_Pain/0ad5a8ed3540d765e3f869b6a7d31b2cde92067450cedf2e08a8d6ac085360ef.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "0ae0dc6a0ed9d54f4282edaec66226324f96e12ee3884ffdbe94a653ce8e374d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "6f7122b6f86b6c5f4f0b032100f91788fc054f07091cb7ab66237c3ce18378f7", "generation_seconds_batch": 23.59774862404447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4612300395965576, "target_z": 2.093749876940174, "ce": 5.895530700683594, "genuineness": 4.331617832183838, "identity": 0.6925768852233887, "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/emo_Pain/0ae0dc6a0ed9d54f4282edaec66226324f96e12ee3884ffdbe94a653ce8e374d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "0b1f1f8f6c28d3baa77ca9eaeadb3a29f5d5c116750899b9743606dbe376f7be", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "b687f8c25f34050d13562ffc69ffb0d71a42625642769750fb7ba847a0039962", "generation_seconds_batch": 20.961014847038314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.245540738105774, "target_z": 1.7187499546621694, "ce": 5.514282703399658, "genuineness": 4.457139015197754, "identity": 0.6126693487167358, "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/emo_Pain/0b1f1f8f6c28d3baa77ca9eaeadb3a29f5d5c116750899b9743606dbe376f7be.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "0b3826d70f24258c548a0ccc5b2e7a92260b713aeb844b5a4ea6fc23e7d72069", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c810fa03a669c531bb1d0d0542dd21d4437469861e0885af9f74793a99089854", "generation_seconds_batch": 23.633160769939423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2137143909931183, "target_z": -0.07519530501116191, "ce": 5.855817794799805, "genuineness": 4.005263805389404, "identity": 0.5619126558303833, "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/emo_Pain/0b3826d70f24258c548a0ccc5b2e7a92260b713aeb844b5a4ea6fc23e7d72069.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "0c200d714bad91fe49790422156e9b092b80229c63797ae15ec481d3a7ded2ee", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e7166623a75676a0c55d2b751c343b84576282284707ce9d50f962ed412c2086", "generation_seconds_batch": 22.52070052397903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3893336057662964, "target_z": 1.968749902847506, "ce": 5.597680568695068, "genuineness": 4.582659721374512, "identity": 0.6289038062095642, "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/emo_Pain/0c200d714bad91fe49790422156e9b092b80229c63797ae15ec481d3a7ded2ee.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "0c23e9f13d47ab065903f9ba532fb9154d546128886b8b6d07dbe438e2f305fb", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "aebc7f7064df9258905385a61088c569f51f3ec0c6408cea290001e0bc4a6378", "generation_seconds_batch": 17.690550341969356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.205295085906982, "genuineness": 4.8588056564331055, "identity": 0.7186568379402161, "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/emo_Pain/0c23e9f13d47ab065903f9ba532fb9154d546128886b8b6d07dbe438e2f305fb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "0c419488400f2b9e9b28fa15f7a8dd25097de008f7e7d94bd9891e65ebb1af98", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "791d28c0ce6ed2e4b192fb74b8747153e8c80aac1005de7dbc11e47aa869e4bb", "generation_seconds_batch": 25.48191570106428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5331264734268188, "target_z": 2.2187498510328423, "ce": 5.673136234283447, "genuineness": 4.105680465698242, "identity": 0.36111894249916077, "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/emo_Pain/0c419488400f2b9e9b28fa15f7a8dd25097de008f7e7d94bd9891e65ebb1af98.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": "0c994bb3b6779f11f4e3122549fd7cacbfb9e4c2420d25d0eae25a75368dc4bf", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8669969e76f8ede48c6459c17ae7df178c4633ee245c8a90babd141a7737e908", "generation_seconds_batch": 17.007002598023973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6366677284240723, "target_z": 0.6601562750977277, "ce": 6.181467056274414, "genuineness": 4.632867813110352, "identity": 0.5444448590278625, "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/emo_Pain/0c994bb3b6779f11f4e3122549fd7cacbfb9e4c2420d25d0eae25a75368dc4bf.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "f32c02856523c67fce41435b07977e7c63fe56d10229c205cde4886f619ca8fd", "generation_seconds_batch": 16.749554385081865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.021054372191429138, "target_z": -0.4101562491903959, "ce": 6.23706579208374, "genuineness": 4.406930446624756, "identity": 0.4563111960887909, "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/emo_Pain/0cb70e86df8754d3bba44fe7f798aac7d0ffe2ca68fc75b92ff1a1ed29986ba7.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": "0d86cf8e985bfcf5973685ef41c0cb7f5291da270490e257303dbee5a44a9981", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "d345f3f32683ab687f435a5cbc968d32165e350ea8749c9319085c9c19f26881", "generation_seconds_batch": 17.2589604689274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1961119174957275, "target_z": 1.6328124336124623, "ce": 5.831989288330078, "genuineness": 4.337893962860107, "identity": 0.8021465539932251, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0d86cf8e985bfcf5973685ef41c0cb7f5291da270490e257303dbee5a44a9981.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": "0db6ae7fd1b34be27b9962116ba8f29dc49edcdfbc8c73b0e3e1a5e0bb1bef18", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "34d8154f343bec7d995de3dba0b90ab39e6feb265c8519f2072aa74a2629a7ad", "generation_seconds_batch": 18.72716961801052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4827643632888794, "target_z": 0.3925780986878662, "ce": 6.221180438995361, "genuineness": 4.482243061065674, "identity": 0.6415830850601196, "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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0db6ae7fd1b34be27b9962116ba8f29dc49edcdfbc8c73b0e3e1a5e0bb1bef18.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "0e1a3b2feec5a8ffb41f4f36c4c27dd50058276afd720153f277973520cc7c18", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "18a28cc06f0675c8c619bff143b7899c338f7d74b9b65aa7942abcc265723959", "generation_seconds_batch": 24.344059766968712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.48388776183128357, "target_z": 0.3945312524288124, "ce": 5.669164657592773, "genuineness": 4.425758361816406, "identity": 0.5193109512329102, "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/emo_Pain/0e1a3b2feec5a8ffb41f4f36c4c27dd50058276afd720153f277973520cc7c18.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": "0e42509580507c588d3f1cb90c2572473443c4279001a1fa68610735b32f186c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "aa1a4ca0366bbe18710c39d2b7dfd11c25311f13f1882fe95c75e9b79b67c6e1", "generation_seconds_batch": 21.642902194987983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646572351455688, "target_z": 1.5781249579005858, "ce": 5.696964263916016, "genuineness": 4.344170093536377, "identity": 0.3659559190273285, "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/emo_Pain/0e42509580507c588d3f1cb90c2572473443c4279001a1fa68610735b32f186c.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": "0e9c66717537923cb1d15c6b34b804a5ff969be83e956786876e9db8166c9c5a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "783646a2b6d607356cdc59b8aa01d944721753cf3bb4a24d2722e84b42bd562e", "generation_seconds_batch": 17.40354120999109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1946168839931488, "target_z": -0.10839845227527514, "ce": 6.046441555023193, "genuineness": 4.519899368286133, "identity": 0.7857778072357178, "hypothesis": "Les clés sont sur la table de de la cuisine. J'ai regardé dans le tiroir, puis près de la fenêtre. Prends-les quand tu partiras.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0e9c66717537923cb1d15c6b34b804a5ff969be83e956786876e9db8166c9c5a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "0edabe4c35d341e7816fead860b0a80b22f6ff3c9cb8ad04f02465b7d98a3845", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "9ae393af4d32abd39783d5f7b128f102bbc2b193e69c9057c9a5e42177e40a98", "generation_seconds_batch": 23.41836972197052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6409711837768555, "target_z": 2.4062499158011716, "ce": 5.035736560821533, "genuineness": 4.005263805389404, "identity": 0.4436330795288086, "hypothesis": "De deur staat nog deur staat nog open. Ik heb jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.21052631578947367, "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/emo_Pain/0edabe4c35d341e7816fead860b0a80b22f6ff3c9cb8ad04f02465b7d98a3845.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "0f1664b9b15adb38d638bfd6756efd10e4240ec47addba291cbfe861ba3ad5d7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "42a9b3da5a930d06c46b001fc49ec541294978b022805707b1ba49dace68d89a", "generation_seconds_batch": 21.989109852002002, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.469283789396286, "target_z": 0.36914061649915675, "ce": 5.272031307220459, "genuineness": 4.118232727050781, "identity": 0.35965394973754883, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschrauben.", "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/emo_Pain/0f1664b9b15adb38d638bfd6756efd10e4240ec47addba291cbfe861ba3ad5d7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "0fa120fd21bc8ad8033c27bf6bea0ef1bd82023a036dd6b3a772570ca8123348", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7246adb6314ec53cd7782367b4bf2deb2d55f954de948c1a2e1a17c53328ffde", "generation_seconds_batch": 22.74051762302406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13563933968544006, "target_z": -0.2109374951423753, "ce": 6.038498878479004, "genuineness": 4.469690799713135, "identity": 0.4532775580883026, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/0fa120fd21bc8ad8033c27bf6bea0ef1bd82023a036dd6b3a772570ca8123348.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "0ff07aec16969ce3710ec4c56bca5310c89264ea4ae0ddf8d903dac242c33df4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c2621cdcc6fb0e29838bfba5999baf2400cf0cd6d4ab7bf34aef832be2714529", "generation_seconds_batch": 23.2788880179869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.221180438995361, "genuineness": 4.65797233581543, "identity": 0.4023497700691223, "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/emo_Pain/0ff07aec16969ce3710ec4c56bca5310c89264ea4ae0ddf8d903dac242c33df4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "105172bcaf0445054ff6e71431f2395d4afd86ac8566fe0e8963023de97b1fa8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8c18c0934cdee713fb927cb5e985b7515e27c580bfffb3127b21b330e8f5c38f", "generation_seconds_batch": 18.72962139698211, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.036781713366508484, "target_z": -0.38281251133445765, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.7383925914764404, "hypothesis": "Ik heb het kleine doosje gevonden, 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/emo_Pain/105172bcaf0445054ff6e71431f2395d4afd86ac8566fe0e8963023de97b1fa8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "1080ed622704caab6cee365e67501d89c0a6103109e3bb1f7786c44ade244587", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ba5e9a4157fe972dbb52cb05e2a455686861f74413fa914053360c412fd3133b", "generation_seconds_batch": 16.381978317978792, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.284721851348877, "genuineness": 4.595211505889893, "identity": 0.7464991211891174, "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/emo_Pain/1080ed622704caab6cee365e67501d89c0a6103109e3bb1f7786c44ade244587.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "10ba6ea452f561c78ddf0fc8e0eb792b7639849f6a6bd56419e7799b0838c9a8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9982078f61cc8979a247fc8c8ae5d2dbd78ccd45bdef97578d667e3d42cbd575", "generation_seconds_batch": 22.979316942975856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5007385015487671, "target_z": 0.42382814402569674, "ce": 5.478540420532227, "genuineness": 4.68307638168335, "identity": 0.562065601348877, "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/emo_Pain/10ba6ea452f561c78ddf0fc8e0eb792b7639849f6a6bd56419e7799b0838c9a8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "11923c0ae34dfcd0f7df1ef63247b89fc719079c3fdfdb442145dfffd051b784", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "eac903ef1dc17d0ea9c5d644c96989f461669f6bfc570d0f9349368acb631d7e", "generation_seconds_batch": 21.792488071951084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6479015350341797, "target_z": 0.6796875016192082, "ce": 5.839932441711426, "genuineness": 4.18099308013916, "identity": 0.5519416332244873, "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/emo_Pain/11923c0ae34dfcd0f7df1ef63247b89fc719079c3fdfdb442145dfffd051b784.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": "124c8ad508fdb95f530c6b1f9c599d3026b306ef786b459107df35fc9ee29c6c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ac629b6ccb88401f67bb75838fe23e706c945138666f29fe8d8d991df56a43b0", "generation_seconds_batch": 16.120575000997633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7512526512145996, "target_z": 0.859374951423753, "ce": 5.831989288330078, "genuineness": 4.105680465698242, "identity": 0.32456839084625244, "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/emo_Pain/124c8ad508fdb95f530c6b1f9c599d3026b306ef786b459107df35fc9ee29c6c.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "0122cfa38ff96b839b35c947dfcc159ddeffb9b421e5c2d6cb7bdf0bf15abf32", "generation_seconds_batch": 16.815602077986114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.23706579208374, "genuineness": 4.3881025314331055, "identity": 0.5631704330444336, "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/emo_Pain/12a7b850c68b1e935d69d2b179f3c0df54c4775061935760902fd9f6cbfc7541.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "13559f28e6ea42e6097c30322da29534d3de24d9f07b8d1ee13328316ed12b1f", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6eacf94b2be9c8d7ca5e600e66066e292a718f8858216c84e41d09f889aa1703", "generation_seconds_batch": 16.82599448703695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08284038305282593, "target_z": -0.3027343737855938, "ce": 6.213237762451172, "genuineness": 4.300237655639648, "identity": 0.34754711389541626, "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/emo_Pain/13559f28e6ea42e6097c30322da29534d3de24d9f07b8d1ee13328316ed12b1f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "13ed767ebd658963035f132076bc2632c0853bd14655bb9a9c51d9594256496b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "44f0123f471cb7fa43be463da5a04465d5639134dc10a6d3aa10bab53de8ed87", "generation_seconds_batch": 23.447729049948975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 6.022613525390625, "genuineness": 4.632867813110352, "identity": 0.5645416975021362, "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/emo_Pain/13ed767ebd658963035f132076bc2632c0853bd14655bb9a9c51d9594256496b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "140a6f03e2d1f16781d134854aa02bc9ab7a867c271ba3b28c18d24f69ab9149", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c83d5c3269842c4ab1f86d5cc774eb643b3ab0e2124e6e7dae0c3733150b44d0", "generation_seconds_batch": 21.966157326009125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 5.911416053771973, "genuineness": 4.419482707977295, "identity": 0.5271166563034058, "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/emo_Pain/140a6f03e2d1f16781d134854aa02bc9ab7a867c271ba3b28c18d24f69ab9149.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": "142d920cdc8e03c5e28ddb084a0a4b9df59a6cf217c2484093992ffc81f3481e", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "de71e87144271f809b2e0d1013a85bf1201a335ac661d673b019d21f2930f859", "generation_seconds_batch": 22.25505174405407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3174371719360352, "target_z": 1.8437499287548378, "ce": 5.871703147888184, "genuineness": 4.256305694580078, "identity": 0.04875211790204048, "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/emo_Pain/142d920cdc8e03c5e28ddb084a0a4b9df59a6cf217c2484093992ffc81f3481e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "14504b3bc9c90c0eb03bbfcc5c5e4f7074ef491139157fc38fe046ddf02fadf4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "658eb70d596efeb5bc83e518d7685360f1ff6bff1cb5d0ffa72700cdad41155b", "generation_seconds_batch": 23.210424395976588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.223073124885559, "target_z": 1.6796875016192083, "ce": 5.6334228515625, "genuineness": 4.494794845581055, "identity": 0.5373419523239136, "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/emo_Pain/14504b3bc9c90c0eb03bbfcc5c5e4f7074ef491139157fc38fe046ddf02fadf4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "14691a04a98706ff1aff6a195606a05fd8dcad099cd95d7e5bc3cc04929c9727", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9f4a28a4df229602b19c66281942c6717a20442e22473e92e8b3bd03bbd7be91", "generation_seconds_batch": 17.256684998050332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8231490850448608, "target_z": 0.9843749255164213, "ce": 6.070269584655762, "genuineness": 4.68307638168335, "identity": 0.7022222280502319, "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/emo_Pain/14691a04a98706ff1aff6a195606a05fd8dcad099cd95d7e5bc3cc04929c9727.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "15214e7f61670aaa923678b671351e1bef0a98b26b7d8299671595ea0bab6862", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f9f80a27d26a55386570c20f253ffaa957e0a817f786b3655002c2559815b5d0", "generation_seconds_batch": 24.08801390195731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4792041778564453, "target_z": 2.124999922278005, "ce": 5.561938762664795, "genuineness": 4.250029563903809, "identity": 0.5293694734573364, "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/emo_Pain/15214e7f61670aaa923678b671351e1bef0a98b26b7d8299671595ea0bab6862.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "153cb16bf1e4d08a739702509ea55df04a4eff0ee2354c93c5bde01b5f3a63e3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6d3c846afd2d91d3e241703ff25c6071e12b77e7686bb5007a2b1ea2ff7d04e4", "generation_seconds_batch": 22.454918506089598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.006728172302246, "genuineness": 4.13078498840332, "identity": 0.8103898763656616, "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/emo_Pain/153cb16bf1e4d08a739702509ea55df04a4eff0ee2354c93c5bde01b5f3a63e3.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "0ef26f524d93be134feed1e9b15658f9cb8a78a05882760772262cee676e30bc", "generation_seconds_batch": 19.595872953068465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.300607204437256, "genuineness": 4.557555198669434, "identity": 0.6127287149429321, "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/emo_Pain/15585f3962d31fd159bb39a755593ab24aa0df87e4725d7772f40811258760f5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "15a89f6e459d8b73eae1cc5de435cec1255f977fab0c506328f4c22d076864e6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "98efb0d760a830ada099a3913a4c9a3fb770ba1c52d17e0d7b3f2da1d0b4c27e", "generation_seconds_batch": 23.51374213292729, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1916184425354004, "target_z": 1.6250000259073318, "ce": 5.458683967590332, "genuineness": 4.143336772918701, "identity": 0.4258662462234497, "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/emo_Pain/15a89f6e459d8b73eae1cc5de435cec1255f977fab0c506328f4c22d076864e6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "15ab4482ac19010ebee7ca5e3c69e5df52702dcd25a22ae90593198ebb5f8f13", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "917a0d2a3a16ff19ecb7d8157fa150cd6ef677a318fac6213301081fef53beae", "generation_seconds_batch": 23.587319644051604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33335456252098083, "target_z": 0.1328124854271259, "ce": 5.530168056488037, "genuineness": 4.532451152801514, "identity": 0.6146065592765808, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem... alten Fenster. Wir können gemeinsam hinein.", "wer": 0.0625, "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/emo_Pain/15ab4482ac19010ebee7ca5e3c69e5df52702dcd25a22ae90593198ebb5f8f13.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "15ad1d741214620e0f1b608cd2e61edc672585eed8b698e5dbe3eae88756ba15", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ca1995825ed5d924435a3e7ebc22267215a57b56156f8ea15058c1019077d708", "generation_seconds_batch": 22.558167097973637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26482826471328735, "target_z": 0.01367186892796912, "ce": 6.205295085906982, "genuineness": 4.532451152801514, "identity": 0.5571659803390503, "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/emo_Pain/15ad1d741214620e0f1b608cd2e61edc672585eed8b698e5dbe3eae88756ba15.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "15ec41af9edf58af8300ba695ca59c135c717eb2b56608c6bd4e0f94e88ccd09", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "620c9da802c921cc94b42407be05aa904c85f1ec2a4357dbc04e6ff224352351", "generation_seconds_batch": 27.175588411046192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0075737833976745605, "target_z": -0.43359375728643706, "ce": 5.903473377227783, "genuineness": 4.319066047668457, "identity": 0.5377415418624878, "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/emo_Pain/15ec41af9edf58af8300ba695ca59c135c717eb2b56608c6bd4e0f94e88ccd09.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "163540e541bc544f8b59cc74b114803e66f7360cb7b3120c3dfcb43d8a72aaf1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "345b61cd67c8f3274a0864e79a1f4746f37ee46f4cbeb24b52383ae2b8c2b933", "generation_seconds_batch": 26.16964961704798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10530802607536316, "target_z": -0.2636718689279691, "ce": 6.316492557525635, "genuineness": 4.13078498840332, "identity": 0.25496089458465576, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/163540e541bc544f8b59cc74b114803e66f7360cb7b3120c3dfcb43d8a72aaf1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "176b46a16531d0586914ddac2ee79e97022c07c45c171f5d18f848d2d048488a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0742b88af2c2590abc3c4a3ab1ef8aa136f970e33b506af75da547247cc36d59", "generation_seconds_batch": 16.93434958695434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 5.95112943649292, "genuineness": 4.545003414154053, "identity": 0.8107534050941467, "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/emo_Pain/176b46a16531d0586914ddac2ee79e97022c07c45c171f5d18f848d2d048488a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "17e4d9eaf0dbaaabfd5418be20532a61c53ca245d9bd900ac761833c4b20adda", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4a3d4eba0791ee1e9774e22f3db4f487f8785c7e397cd7a363bb04a7d7f41104", "generation_seconds_batch": 24.054555406910367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1755194067955017, "target_z": -0.14160154772472486, "ce": 5.752562999725342, "genuineness": 4.482243061065674, "identity": 0.6850659251213074, "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/emo_Pain/17e4d9eaf0dbaaabfd5418be20532a61c53ca245d9bd900ac761833c4b20adda.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "1852f7baa86108e9b9cd2a5f3dfc260c36c0430fe66c8fc1d0e261c84f6f307e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "128922cbf60b41485ca9e8e25299b058f43345b988ef713bb88003b6e29b2b64", "generation_seconds_batch": 24.110118541982956, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11204831302165985, "target_z": -0.2519531278336144, "ce": 6.300607204437256, "genuineness": 4.770940780639648, "identity": 0.501490592956543, "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/emo_Pain/1852f7baa86108e9b9cd2a5f3dfc260c36c0430fe66c8fc1d0e261c84f6f307e.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": "18536624586f28c84008239f3055e9e33ef4014ce86a935140443fe491b62272", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f772e81c595a0a55dc8376dd20285ae6121a1452c9a3faac8ce79873b54aa89", "generation_seconds_batch": 15.661752338055521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29459789395332336, "target_z": 0.06542969175042161, "ce": 6.292664527893066, "genuineness": 4.7081804275512695, "identity": 0.7443852424621582, "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/emo_Pain/18536624586f28c84008239f3055e9e33ef4014ce86a935140443fe491b62272.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "18879ac6216eeb799bd5523a548c26b7bce59709ef3e5a7cb2a775dc7c632a20", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ef0830a20424b9a1b6269aa50994219277223522faddc6c8c53861a51d3e99e", "generation_seconds_batch": 16.560726227005944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08284038305282593, "target_z": -0.3027343737855938, "ce": 6.213237762451172, "genuineness": 4.68307638168335, "identity": 0.8269321322441101, "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/emo_Pain/18879ac6216eeb799bd5523a548c26b7bce59709ef3e5a7cb2a775dc7c632a20.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": "18a775ad5379ec1473b50f1244cca18464c706a1986d231ff05dfbb9a0897dce", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "852fcf8259cb2701a4749ab71f93c39baed82f7bd4fce1777c20ed9be55714a8", "generation_seconds_batch": 25.231766785960644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8276426792144775, "target_z": 0.9921875404802059, "ce": 5.685050010681152, "genuineness": 4.607763767242432, "identity": 0.7892364859580994, "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/emo_Pain/18a775ad5379ec1473b50f1244cca18464c706a1986d231ff05dfbb9a0897dce.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "18e2fd2e1a6680652731af9462267f28902f18a6f42d07c6009ca0117d7c1b4f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "144e43cf62fce2386abf2d74f2e03470a1a1c01cbf7c66ef5d670894f9335e23", "generation_seconds_batch": 22.74051762302406, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 5.998785495758057, "genuineness": 4.457139015197754, "identity": 0.36293309926986694, "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/emo_Pain/18e2fd2e1a6680652731af9462267f28902f18a6f42d07c6009ca0117d7c1b4f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "18e67db22990845919bbed03198a5f606df24ee35a74ba70f51abdc79f701653", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "50445e1a542ac78eadc6ee916dbd6f09ccdb644d9dc95e14a1e4d05935d326ab", "generation_seconds_batch": 22.96319531800691, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9399808645248413, "target_z": 1.187500012953666, "ce": 5.355429172515869, "genuineness": 4.105680465698242, "identity": 0.5844219326972961, "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/emo_Pain/18e67db22990845919bbed03198a5f606df24ee35a74ba70f51abdc79f701653.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "18f50bc6c6b4e5dadde88d49ea7643983262f65fd182c2661af5f982115acfde", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9c3aab527d1f9b81cc184224c0bddd986234339f08910994dd8013e576586fea", "generation_seconds_batch": 22.738668750040233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08508715033531189, "target_z": -0.298828118118365, "ce": 6.23706579208374, "genuineness": 4.65797233581543, "identity": 0.29053422808647156, "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/emo_Pain/18f50bc6c6b4e5dadde88d49ea7643983262f65fd182c2661af5f982115acfde.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "190e0b4e812b4994e8c135f4d19723131b89e35d0650a1ed4d163e48eb952226", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f53595c4c4fb9d18d63cc9a85c1629ad10b8d02a7d3cb7d0b680f8a697f37048", "generation_seconds_batch": 23.821633242070675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24573077261447906, "target_z": -0.019531252428812353, "ce": 6.1020402908325195, "genuineness": 4.400654315948486, "identity": 0.4151008129119873, "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/emo_Pain/190e0b4e812b4994e8c135f4d19723131b89e35d0650a1ed4d163e48eb952226.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "19502ff6da74666ac049b1730ffb2b7429b9afcb40a63f76511528083d26d420", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6892c7b3747cb269ff50d8bc08034d4c13cdb9fcb59c282c68802386a2ad487f", "generation_seconds_batch": 19.901732868980616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.800681471824646, "target_z": 0.9453124724734601, "ce": 5.728734970092773, "genuineness": 4.362998008728027, "identity": 0.3971642255783081, "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/emo_Pain/19502ff6da74666ac049b1730ffb2b7429b9afcb40a63f76511528083d26d420.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "195343b6a2d280c6e370db09308a5c28f6b9059ef91eb5fa5a3ca85ac067c5c4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fb51b8ed3ef3bd7c34ea4cc0ca0d3bdd98fcbd23e1cf4c05a8f5137582d15847", "generation_seconds_batch": 17.56555771001149, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05138568580150604, "target_z": -0.3574218754048021, "ce": 6.332377910614014, "genuineness": 4.6203155517578125, "identity": 0.4411424398422241, "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/emo_Pain/195343b6a2d280c6e370db09308a5c28f6b9059ef91eb5fa5a3ca85ac067c5c4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "1b72d0a0c8a584b8a45c057ddf5a6b26c393c00498507c92693eb6b28e3a1ea4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bee180a703e5c95306e23d89f81fcba154d4e291f667f539581438bd1a298f1d", "generation_seconds_batch": 23.55150776996743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07272994518280029, "target_z": -0.3203124983807918, "ce": 6.022613525390625, "genuineness": 4.532451152801514, "identity": 0.6370111703872681, "hypothesis": "Ma porte est encore ouverte. 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/manual-comparison-v1/audio/emo_Pain/1b72d0a0c8a584b8a45c057ddf5a6b26c393c00498507c92693eb6b28e3a1ea4.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": "1c6b2af7661e70e9dfb90a0a9169ea41b3ec232623fcf2bc7a629d62a0706a01", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "21c148fca7ffd3eed592425d65a200589628737922b6a5de43c6762d54efdbd5", "generation_seconds_batch": 17.972136022988707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.140132874250412, "target_z": -0.20312498380791766, "ce": 5.816103935241699, "genuineness": 4.532451152801514, "identity": 0.7471289038658142, "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/emo_Pain/1c6b2af7661e70e9dfb90a0a9169ea41b3ec232623fcf2bc7a629d62a0706a01.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "1ca6fba11d6fd36d89d466caebd76bf07c11557830fecbe1a79b9b53717bd1f1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9532970bfc7989057c56435f3c97383be1c720d49b4e66e28bd1482414215174", "generation_seconds_batch": 24.021651820046827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.229123115539551, "genuineness": 4.375550270080566, "identity": 0.6498533487319946, "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/emo_Pain/1ca6fba11d6fd36d89d466caebd76bf07c11557830fecbe1a79b9b53717bd1f1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "1cf16abd260a58c4e6b5c1a999fd7c798c85e8e3e107971ec78e58a878d3c98d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "02c32c022c9de66fce1ae021ee47099dc203840e8337862457f47678e82f1857", "generation_seconds_batch": 22.94222763599828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8207123279571533, "target_z": 2.7187499546621696, "ce": 5.879645347595215, "genuineness": 4.024092197418213, "identity": 0.7257524728775024, "hypothesis": "The door is still open. Put your coat on the chair. Tell me when you are ready", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/1cf16abd260a58c4e6b5c1a999fd7c798c85e8e3e107971ec78e58a878d3c98d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "1d10ef83f6f9d3629abe7403a4e66d611cefc426bcf8508ed1123acd7086b8eb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c709ceeb63377aca1c393163bfdb4b02a795d1d1e968f5b5a131487890347a07", "generation_seconds_batch": 18.53094234294258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2856108248233795, "target_z": 0.04980466908150633, "ce": 5.577824115753174, "genuineness": 4.432034492492676, "identity": 0.7109711170196533, "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/emo_Pain/1d10ef83f6f9d3629abe7403a4e66d611cefc426bcf8508ed1123acd7086b8eb.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "36b275826c32c0492433f8390093ec192c9973542fc000baa544cad3c806d7d1", "generation_seconds_batch": 16.623498575994745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0480155348777771, "target_z": -0.3632812589056453, "ce": 6.300607204437256, "genuineness": 4.413206577301025, "identity": 0.6855944991111755, "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/emo_Pain/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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "b0995e0a9d9ebdc05f072c2f0ac2a9603d4332586e668832e45d3a26406675c1", "generation_seconds_batch": 16.269342013052665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.1894097328186035, "genuineness": 4.532451152801514, "identity": 0.7495160698890686, "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/emo_Pain/1d829402ab2176d2024ae3e918286a39ec094d9a5299caa0e048ba37912e1579.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "56a9123732a1089ca7bc3d624d2067e733a7ff42582c7a9d216f526f78386f30", "generation_seconds_batch": 16.623498575994745, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04015186429023743, "target_z": -0.3769531278336144, "ce": 6.292664527893066, "genuineness": 4.1998209953308105, "identity": 0.5410726070404053, "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/emo_Pain/1eaf0e0e2c966186cf6ada2de41819a315f8b1b1eeeeef5a84a28dd582998db8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "1eceedfd8763e9cd93d3fa23b2ae7a3067a3e0a236659732dafa08473316c9bc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "85e3acfb25e859596a253a32ffcb9f0ec979720b2c4da6f93b5e454af4ac2d33", "generation_seconds_batch": 18.326504791039042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 5.808161735534668, "genuineness": 4.532451152801514, "identity": 0.7261567711830139, "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/emo_Pain/1eceedfd8763e9cd93d3fa23b2ae7a3067a3e0a236659732dafa08473316c9bc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "1f0a56377dce40abeffb5ba1c66a53219f4a077c18669a9a8c88932684d11e35", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "268a4d6fedc9723b2206a8546953e475cc921242f6f76c40ba7ae18989140553", "generation_seconds_batch": 21.951129827066325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016560837626457214, "target_z": -0.41796876052485354, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.609530508518219, "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/emo_Pain/1f0a56377dce40abeffb5ba1c66a53219f4a077c18669a9a8c88932684d11e35.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": "1f6e58811f092b9a422969a7d576f7ae694efcd4eec597e2c5968ee90ffbf57a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "425d1f0f76e673b6a4ae2e77c46f37ee766784553efaf1fe59fb296a17b1adae", "generation_seconds_batch": 23.057104796054773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3103252351284027, "target_z": 0.09277342960635986, "ce": 5.689021587371826, "genuineness": 4.337893962860107, "identity": 0.7168264985084534, "hypothesis": "", "wer": 1.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/1f6e58811f092b9a422969a7d576f7ae694efcd4eec597e2c5968ee90ffbf57a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "20b586a3f793595bb06284ca446e6811a9b9d2ebc6359c13a1af14a2aee5aa80", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2dcfe02fcb7dc14c92684efc99371701cde9aaea767b81b46807b31f191cb7ac", "generation_seconds_batch": 22.36974727199413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16147711873054504, "target_z": -0.16601563269123912, "ce": 5.736677646636963, "genuineness": 4.093128204345703, "identity": 0.5399924516677856, "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/emo_Pain/20b586a3f793595bb06284ca446e6811a9b9d2ebc6359c13a1af14a2aee5aa80.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": "2164db25d9b768e088d0dda722ff2ec536515924cb672c8647ab0a081ea2d204", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "0bc24d2f1266cd3bb69492dc063b66418c3848e58571d7c38f7ddeb0e8deb695", "generation_seconds_batch": 17.241347386036068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10193787515163422, "target_z": -0.2695312524288124, "ce": 6.070269584655762, "genuineness": 4.808597087860107, "identity": 0.4940565228462219, "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": 0.120000000000001, "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/emo_Pain/2164db25d9b768e088d0dda722ff2ec536515924cb672c8647ab0a081ea2d204.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "21b9b4ddc8d5154ee7dd4d5e838e9c8aebe442b5bc33ac6ab570fc0cbe02d222", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "86de4b0acfbc13be07a60333c93387a3e7a131c476050d483a937e066dade4b8", "generation_seconds_batch": 18.172874436015263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38671523332595825, "target_z": 0.22558596684814924, "ce": 6.038498878479004, "genuineness": 4.444586753845215, "identity": 0.7300312519073486, "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/emo_Pain/21b9b4ddc8d5154ee7dd4d5e838e9c8aebe442b5bc33ac6ab570fc0cbe02d222.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": "2294e55cd45bc6086bd49e3658b4bc6edb9c02607dd36099560da2030092240b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "de80e71b8e1040428be7dd0cd51ceb942c28681ebd915892c4aec177cc769da5", "generation_seconds_batch": 17.38699952198658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28448745608329773, "target_z": 0.04785156715522367, "ce": 6.022613525390625, "genuineness": 4.356721878051758, "identity": 0.22760289907455444, "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/emo_Pain/2294e55cd45bc6086bd49e3658b4bc6edb9c02607dd36099560da2030092240b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "2298cd4c09d1d81015bbbf4fd0d7e78eb96b0553dd8a438e1da1fd7ca67cfad4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "28471411dc12ea8f7658415ed08a29c4049d76bdef85e3b23c7dd25088f4f499", "generation_seconds_batch": 22.52070052397903, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2949695587158203, "target_z": 1.8046874757118765, "ce": 5.061550617218018, "genuineness": 4.168440818786621, "identity": 0.6999892592430115, "hypothesis": "La porte la porte est encore ouverte j'ai poussé ton manteau sur la chaise dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/2298cd4c09d1d81015bbbf4fd0d7e78eb96b0553dd8a438e1da1fd7ca67cfad4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "242567dfbd96a40ae9bf4cc73c7408e6a78e934fff4976139695c3383bf4363a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dc6a4188e5e34418fab97513be15de87796d79356f0cb2d123425ac0e37db719", "generation_seconds_batch": 16.96423662104644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7467591762542725, "target_z": 0.8515625437186224, "ce": 6.117925643920898, "genuineness": 4.250029563903809, "identity": 0.4857645332813263, "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/emo_Pain/242567dfbd96a40ae9bf4cc73c7408e6a78e934fff4976139695c3383bf4363a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "242d0da0dad1e0124db119042ba57ab03aa60d15740dc693a64c610b2fd7eb36", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "69c2ce8e308e801fe82d82407dde0fa9c256649088b0cd03ad038cf8dc116a78", "generation_seconds_batch": 24.344059766968712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8298894166946411, "target_z": 0.9960937443327712, "ce": 5.637393951416016, "genuineness": 4.444586753845215, "identity": 0.5933660268783569, "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/emo_Pain/242d0da0dad1e0124db119042ba57ab03aa60d15740dc693a64c610b2fd7eb36.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "242efd9590a64c69d32810f16415605e8128c23f87d661d113c937263589e410", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "de08d5ad795016dd4e86522e4ff1d0d2b1732afd046d8fad304a9d9fee4b7db0", "generation_seconds_batch": 26.7373935719952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 6.395919322967529, "genuineness": 4.7081804275512695, "identity": 0.6853927373886108, "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/emo_Pain/242efd9590a64c69d32810f16415605e8128c23f87d661d113c937263589e410.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "2463b6f9215d841659f1b262ba8a8e6aa5c85b5fbc139f5531298cf65351923b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "cf598e49c13977d840c20cbcedf82de90a2da3b302b8783591eff3620f7f372e", "generation_seconds_batch": 16.96423662104644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8770714998245239, "target_z": 1.078125061529913, "ce": 6.070269584655762, "genuineness": 4.400654315948486, "identity": 0.45578694343566895, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/2463b6f9215d841659f1b262ba8a8e6aa5c85b5fbc139f5531298cf65351923b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "246d56ede52999719b5e6f485b8aa4fc89bf45d79815cd0e11055ed4c2c260ff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b41c9b70e45323c76e5fef4f472397fe3cc252bf8059252ecdd9624cf96b5bab", "generation_seconds_batch": 22.397373902029358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12159706652164459, "target_z": -0.2353515542015578, "ce": 5.8875885009765625, "genuineness": 4.896461486816406, "identity": 0.6769863367080688, "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/emo_Pain/246d56ede52999719b5e6f485b8aa4fc89bf45d79815cd0e11055ed4c2c260ff.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "248ad9094975aabf2e22a00a54cf9068a7b912a02f9badf4856982e1f034b449", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "599975c02c89faf8644c5aa4b1f4ceb067fe1667a46222bdd6783e7781245595", "generation_seconds_batch": 22.469890577020124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3794132471084595, "target_z": 0.21289064888332146, "ce": 5.546053409576416, "genuineness": 4.532451152801514, "identity": 0.6188759803771973, "hypothesis": "que he 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/emo_Pain/248ad9094975aabf2e22a00a54cf9068a7b912a02f9badf4856982e1f034b449.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "24acae6e7d6174912a26f067b6d7489cefc2b0ed3ad4cb069dff276e11936097", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d97e13fc59d21c4f1386adddb6a56b33d2f9d19f3695dedb7907732b8a6b684d", "generation_seconds_batch": 21.367333467002027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16147711873054504, "target_z": -0.16601563269123912, "ce": 5.959072113037109, "genuineness": 4.344170093536377, "identity": 0.6274679899215698, "hypothesis": "De deur staat nog open. Ik heb 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/emo_Pain/24acae6e7d6174912a26f067b6d7489cefc2b0ed3ad4cb069dff276e11936097.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": "24e4520e3a0b74a1c3a124451562f1d77fccd060c44ade44326e6e37d196f399", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "06307461050981443a104e25c888dfabb80192917577d80e97be6e57d1ea2644", "generation_seconds_batch": 23.474090890958905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 5.784333229064941, "genuineness": 4.595211505889893, "identity": 0.6384822130203247, "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/emo_Pain/24e4520e3a0b74a1c3a124451562f1d77fccd060c44ade44326e6e37d196f399.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "259be954fe832c3015ad34f49b2f8c8c404348c79597343a1acba4444a23465e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f4e45175347a822c44a325a43176339487824be70792aacdd51d1bb3b1fac883", "generation_seconds_batch": 24.208149040932767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19995296001434326, "target_z": -0.09912108858877375, "ce": 5.49045467376709, "genuineness": 4.494794845581055, "identity": 0.7003024220466614, "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/emo_Pain/259be954fe832c3015ad34f49b2f8c8c404348c79597343a1acba4444a23465e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "261ab7c93342f9abcebeb942fbe64f14cf8249fbdb9be7932edaf155bf420770", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a5def0c519d87f835935085c73bf3756fcf15738e7040ad3a57ea4e42549a82d", "generation_seconds_batch": 17.55001882102806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13226918876171112, "target_z": -0.21679687864321853, "ce": 5.80021858215332, "genuineness": 4.645420074462891, "identity": 0.6706552505493164, "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/emo_Pain/261ab7c93342f9abcebeb942fbe64f14cf8249fbdb9be7932edaf155bf420770.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": "2647b55029c112b0e356984835db2e7587769bad7fbd9ab3bd06ef04ac4785c0", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7193aee7210625704cc5d4e25c8b5bb4d3a4285376f5a6c3921e81c5d98dbf35", "generation_seconds_batch": 17.663815699983388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.219331294298172, "target_z": -0.06542969175042161, "ce": 5.967014789581299, "genuineness": 4.20609712600708, "identity": 0.5755867958068848, "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/emo_Pain/2647b55029c112b0e356984835db2e7587769bad7fbd9ab3bd06ef04ac4785c0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "26613ff0a427e1d894bca2382583ed112899c9be1d06c937b175546af87a0167", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "462aaf432a2b01ca6ddddc9df6d5af00b2a69bbe4bff771fb787675747477c4d", "generation_seconds_batch": 23.359805307001807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4342689514160156, "target_z": 2.0468750161920823, "ce": 5.40705680847168, "genuineness": 4.2751336097717285, "identity": 0.6417127847671509, "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/emo_Pain/26613ff0a427e1d894bca2382583ed112899c9be1d06c937b175546af87a0167.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "cdef26cd70fdaf0773dc79bdf8830c8623142acc9ffcbba8e283ed03c2051632", "generation_seconds_batch": 17.76765338901896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 6.467403411865234, "genuineness": 4.545003414154053, "identity": 0.1444093883037567, "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/emo_Pain/26d12995a5b024b42cbe037fdee48ffb67b79df656c09f059592185aa881aca9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "26f077dffaa8585687fd2bc6ebb8c00f2f0e8c3ebcedff89b16b9ab87ef36123", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ad27e626985ccb6e2d8ca22d99e984eb0a4fd83c0ac2eae8cad975ef9df30b25", "generation_seconds_batch": 22.416539905942045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10755479335784912, "target_z": -0.2597656132607403, "ce": 5.673136234283447, "genuineness": 4.469690799713135, "identity": 0.5227116942405701, "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/emo_Pain/26f077dffaa8585687fd2bc6ebb8c00f2f0e8c3ebcedff89b16b9ab87ef36123.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": "27e985447fe2c3dce6c603196e8ddee33877235c22574181834b1e2cd2320ff4", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6cf9aaf3e4cd5c1819392e1e35e74be6fc2cc50f8ee8f5a4633c50ef8272cbb0", "generation_seconds_batch": 17.404119281913154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7759671211242676, "target_z": 0.9023438155779335, "ce": 5.935244083404541, "genuineness": 4.6956281661987305, "identity": 0.5996224880218506, "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/emo_Pain/27e985447fe2c3dce6c603196e8ddee33877235c22574181834b1e2cd2320ff4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "2840202a5349b414f0a0adb48b75609f748849ab268af168bd45d8da95334161", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b8c6a91e11b654b9c315d6858d27410d76ff1ff4b0a29b441105484036d0070d", "generation_seconds_batch": 21.689895846066065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43445897102355957, "target_z": 0.3085937831937688, "ce": 6.125868320465088, "genuineness": 4.570107460021973, "identity": 0.2746885418891907, "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/emo_Pain/2840202a5349b414f0a0adb48b75609f748849ab268af168bd45d8da95334161.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "288c6d978ac26a6045302fe081503eb55287a75b590ad6894ecd5558b49976d9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "12de565c29a2176a96b28fe7f29dd1115895da5c08f7bd17adbb354b34e02bc1", "generation_seconds_batch": 22.601961210020818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08396376669406891, "target_z": -0.3007812459519794, "ce": 6.070269584655762, "genuineness": 4.68307638168335, "identity": 0.6189227104187012, "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/emo_Pain/288c6d978ac26a6045302fe081503eb55287a75b590ad6894ecd5558b49976d9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "293337a20425b2900023c69bc7f82cf224d6761c4eb29d30302cad8c6f4cee08", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "a94f49ba522289908fc01de736b4549cf9d637b6a297bad7a0fb5a1e731e395e", "generation_seconds_batch": 22.95724389597308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8770714998245239, "target_z": 1.078125061529913, "ce": 5.561938762664795, "genuineness": 4.337893962860107, "identity": 0.5130821466445923, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle comprend.", "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/emo_Pain/293337a20425b2900023c69bc7f82cf224d6761c4eb29d30302cad8c6f4cee08.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "294ab376eb584fbcadc64dbce4a23b58033283e5c00ea0a742cbd77592112033", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a561bdb8e265830ccb539a0df1d87707bc57e195520f66dbf03f6800e02d5d1f", "generation_seconds_batch": 23.636065913946368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5421135425567627, "target_z": 2.234374873701758, "ce": 5.911416053771973, "genuineness": 4.287685871124268, "identity": -0.022394021973013878, "hypothesis": "Ik heb het kleine toosje 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/emo_Pain/294ab376eb584fbcadc64dbce4a23b58033283e5c00ea0a742cbd77592112033.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "2965606866091735bd06af4e73dc1c297760c548dd68ca27be1a31305caf35c2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b056d7b073e1bbf294e01a1e31bf783c0dfdfb595ce85986e4b04935cd10e0e2", "generation_seconds_batch": 22.790873651974835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8770714998245239, "target_z": 1.078125061529913, "ce": 5.9431867599487305, "genuineness": 4.607763767242432, "identity": 0.755297839641571, "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/emo_Pain/2965606866091735bd06af4e73dc1c297760c548dd68ca27be1a31305caf35c2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "29cff4cdd1164c3bffa7a4014817032cb8d8f578800faf9175c59968020dc56e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "17ea9363273739f8ce61aa243d5920091380af90cd28c52e8d94c6d2df47169f", "generation_seconds_batch": 20.550634487066418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10193787515163422, "target_z": -0.2695312524288124, "ce": 5.990842819213867, "genuineness": 4.7458367347717285, "identity": 0.6605180501937866, "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/emo_Pain/29cff4cdd1164c3bffa7a4014817032cb8d8f578800faf9175c59968020dc56e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "2a0538053bc6e78a0c6145e4d27918025a8f2563d891650d4a783e831bc6d32c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3aafc591e97ad820ad0f93f6cf75ebfeeba8049ea6c7cc166b07ed700baafaf6", "generation_seconds_batch": 23.064629781991243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 6.157639026641846, "genuineness": 4.369274139404297, "identity": 0.6468141674995422, "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/emo_Pain/2a0538053bc6e78a0c6145e4d27918025a8f2563d891650d4a783e831bc6d32c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "2a26f1ed68be362b09a90c60710d4a8b3ffa7a8f9a3126957583dc4bd08e36e7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2cef1eb97cf0b1f48ae9937bf7ebd7fb5cb2271c275cf80c57fd8314b7b264d4", "generation_seconds_batch": 16.45040664600674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08059361577033997, "target_z": -0.30664062945282267, "ce": 6.09409761428833, "genuineness": 4.6705241203308105, "identity": 0.6417946815490723, "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/emo_Pain/2a26f1ed68be362b09a90c60710d4a8b3ffa7a8f9a3126957583dc4bd08e36e7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "2a469a4cc65d457b96803092c32b31d2e860e7238c4f9ccddd161f660e7096c7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "13f3dd9d9ec062cebaec9abe03bb7f3fa5a89a3c90437a1f578d463dfaf52a3a", "generation_seconds_batch": 21.84408919804264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20219972729682922, "target_z": -0.09521483292154494, "ce": 5.935244083404541, "genuineness": 4.808597087860107, "identity": 0.7015883326530457, "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/emo_Pain/2a469a4cc65d457b96803092c32b31d2e860e7238c4f9ccddd161f660e7096c7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "2a63b6454403e9308ebda6396563b34c3468b080584f50c703a4cc6585d15638", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a09d20c6b7977d191a94a00d816c8bbe7844688e47582954c26702e132796ce0", "generation_seconds_batch": 22.761081618024036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.668122410774231, "target_z": 0.7148437508096042, "ce": 5.911416053771973, "genuineness": 4.6956281661987305, "identity": 0.8234797716140747, "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/emo_Pain/2a63b6454403e9308ebda6396563b34c3468b080584f50c703a4cc6585d15638.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "2b05cab5011680cf16a0d46700d2c9f704a1d191a164b73c95e13190055427f4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a3983a8e87c9f8362889e55bc36b1cc6a5383b97a869b53db39643f5a25b4dc1", "generation_seconds_batch": 21.652912486926652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 6.157639026641846, "genuineness": 4.494794845581055, "identity": 0.3015890121459961, "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/emo_Pain/2b05cab5011680cf16a0d46700d2c9f704a1d191a164b73c95e13190055427f4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "2b2b3e4bcebb3b8e4052348ede3cb0c7ad8a38f027da92f1557f0526d55e613b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "946b0a1f965a4562c388755d29e00f722f1f3ff94d1d680f7577af746487ca8c", "generation_seconds_batch": 17.51046027895063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2904759645462036, "target_z": 1.7968748607480918, "ce": 5.788304805755615, "genuineness": 4.545003414154053, "identity": 0.49004077911376953, "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/emo_Pain/2b2b3e4bcebb3b8e4052348ede3cb0c7ad8a38f027da92f1557f0526d55e613b.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": "2be01da87e5d3ab252be6a702d30663f72e0ab559a0a8cc5f4c4a5b6e6a49d5f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a00617752f7895abe07fecfe09efb296071316acf43f92827f399b742da37c7f", "generation_seconds_batch": 21.43925783596933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15080499649047852, "target_z": -0.18457030824957837, "ce": 5.696964263916016, "genuineness": 4.425758361816406, "identity": 0.597755491733551, "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/emo_Pain/2be01da87e5d3ab252be6a702d30663f72e0ab559a0a8cc5f4c4a5b6e6a49d5f.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": "2c13c5eba545bbcb44e91978283e9e8b83a9d781ac14029b75236912d189186b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5300a9b0a91b65f03b446101d6caa114266d2d6777f6e5117d9535fbfec38441", "generation_seconds_batch": 16.253656171029434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3039566278457642, "target_z": 1.8203124983807917, "ce": 5.935244083404541, "genuineness": 4.268857479095459, "identity": 0.5952677726745605, "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/emo_Pain/2c13c5eba545bbcb44e91978283e9e8b83a9d781ac14029b75236912d189186b.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": "2c2f1d7180f0137dd73641edb0a8dc061340d44ed8fe6aef56bc39596c6f22c9", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "33084dd2d0d4818ff9b3e8c4908a8e122c09c42c244f024431acc0430be1cef7", "generation_seconds_batch": 16.672387900063768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7085641622543335, "target_z": 0.7851562491903958, "ce": 6.181467056274414, "genuineness": 4.7834930419921875, "identity": 0.7898390293121338, "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/emo_Pain/2c2f1d7180f0137dd73641edb0a8dc061340d44ed8fe6aef56bc39596c6f22c9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "2c4cd9c9264f6614a23079baa340267b56f1c3c3410b830691470afd4a2d1352", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5061deb51ee6abc44c9ca2bf57b2ffc7d042146b187497f594e48980cfbf6a62", "generation_seconds_batch": 15.748068358981982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11991198360919952, "target_z": -0.23828125890564528, "ce": 5.895530700683594, "genuineness": 4.494794845581055, "identity": 0.6127972602844238, "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/emo_Pain/2c4cd9c9264f6614a23079baa340267b56f1c3c3410b830691470afd4a2d1352.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "2c5c8391f41b437389a7f7c85ca169e2d515049dd195b46b9b5e552a13fbf212", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3eb7f0837348ef3a3dfed9279ee3b216097d01d3516e2587d3a3c8c1fb4e89e8", "generation_seconds_batch": 23.41836972197052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5241395235061646, "target_z": 2.2031250356225813, "ce": 5.863759994506836, "genuineness": 4.256305694580078, "identity": -0.008153067901730537, "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/emo_Pain/2c5c8391f41b437389a7f7c85ca169e2d515049dd195b46b9b5e552a13fbf212.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": "2d0b211824363b81470aab5d44af14d1147a9e86875e2451931d5de8e88aa689", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f4f49703c10bec6de35150484c9568a13de91c3fa357340930af765e2af0f3d2", "generation_seconds_batch": 22.557077716104686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7669800519943237, "target_z": 0.8867187929090182, "ce": 5.816103935241699, "genuineness": 3.9173994064331055, "identity": 0.6763234734535217, "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/emo_Pain/2d0b211824363b81470aab5d44af14d1147a9e86875e2451931d5de8e88aa689.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": "2e84834658e5aa4c7a6daf141e0feec4e337a21ce4c5bdbee55de9f493482f07", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "75d0e791e25d6d3122e4f846890917e9f833388a9b7b3b4f0acc2e4de654927d", "generation_seconds_batch": 17.530481061083265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2541561424732208, "target_z": -0.004882806630370149, "ce": 5.935244083404541, "genuineness": 4.356721878051758, "identity": 0.7926531434059143, "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/emo_Pain/2e84834658e5aa4c7a6daf141e0feec4e337a21ce4c5bdbee55de9f493482f07.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": "2ec2d7496f22965e7aeeb55d6fda733acf70d852e40ceae1cd84a9287894dd12", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "e8da9059d7c44bb24fc883fcf2d21e2695a80bc15f9ecf1ca2299d6dd849fe17", "generation_seconds_batch": 18.676699334057048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5333166122436523, "target_z": 0.4804688252931829, "ce": 6.157639026641846, "genuineness": 4.20609712600708, "identity": 0.21610988676548004, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de laag en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.045454545454545456, "duration_error_s": 0.120000000000001, "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/emo_Pain/2ec2d7496f22965e7aeeb55d6fda733acf70d852e40ceae1cd84a9287894dd12.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "309f9082bf30e2d8911993d88accaadce73a476647302787e61b7736b9b9cd69", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "2a074b2b66294a6354dff1d0930ba2ac627d381029f0c62eefe49739ac24500c", "generation_seconds_batch": 21.80773731006775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04127524793148041, "target_z": -0.375, "ce": 5.816103935241699, "genuineness": 4.268857479095459, "identity": 0.5582218170166016, "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": 0.120000000000001, "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/emo_Pain/309f9082bf30e2d8911993d88accaadce73a476647302787e61b7736b9b9cd69.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "30b5df301fae7e9198d63c4080b48a4063fb3a1b843649880cf66b5cb070f13e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "673eb34a70e33025a363f8e7a227a496482c8c49d9c6690810bb6dbe5e2cf894", "generation_seconds_batch": 23.381369772017933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5692647695541382, "target_z": 0.54296870871019, "ce": 5.911416053771973, "genuineness": 4.65797233581543, "identity": 0.6816741228103638, "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/emo_Pain/30b5df301fae7e9198d63c4080b48a4063fb3a1b843649880cf66b5cb070f13e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "30c249c1980d98ffd7665751513a7d4e60756168b1e99a4c7d1784ae43b82d5c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "500e409da1ab9d3d89c5163e8a4ec666f195f54e188702d9de2128e0bd0fae50", "generation_seconds_batch": 20.961014847038314, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5344399213790894, "target_z": 0.4824218235901386, "ce": 5.808161735534668, "genuineness": 4.482243061065674, "identity": 0.6694760918617249, "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/emo_Pain/30c249c1980d98ffd7665751513a7d4e60756168b1e99a4c7d1784ae43b82d5c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "30e352a8918c70d7aec64254668874e59c10087cc425978f587bd9af2b4cb5f4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "274b1bdf5b0df11cfe5a96bb995577a0c3f20e193b5d2f320d8bcb1bc711a7bc", "generation_seconds_batch": 17.146380318910815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.203323096036911, "target_z": -0.09326173099526228, "ce": 6.316492557525635, "genuineness": 4.645420074462891, "identity": 0.9078742265701294, "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/emo_Pain/30e352a8918c70d7aec64254668874e59c10087cc425978f587bd9af2b4cb5f4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "314ba4cd8552268359e89827d44e8dda00b10a3464c2351ef35ef22733396148", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "316b0f89b198a6deaadd9fa82c652bf13c2eb8e7d2d84334f65f52e4cf492c9a", "generation_seconds_batch": 21.42293769295793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12047368288040161, "target_z": -0.23730468203517222, "ce": 5.839932441711426, "genuineness": 4.519899368286133, "identity": 0.6615171432495117, "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/emo_Pain/314ba4cd8552268359e89827d44e8dda00b10a3464c2351ef35ef22733396148.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "31ce5668bc137946e08587a4820fe10bb8afa4425c62fc6ac0a2645f5c8b3e1a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c7a4a60a245e81da5370dec78d2bfee0b05baccfe7a26622f1a764a486f7a47d", "generation_seconds_batch": 23.064629781991243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0064504146575927734, "target_z": -0.43554685921271974, "ce": 5.967014789581299, "genuineness": 4.20609712600708, "identity": 0.623813271522522, "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/emo_Pain/31ce5668bc137946e08587a4820fe10bb8afa4425c62fc6ac0a2645f5c8b3e1a.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": "31fedc6248c3c7cb7fecba1d6b9a9fef2ce84d55bcb95459c023484889f83669", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e08b55ddc22824f88ea65a33259f3cac87a3edcf12bd8c45625ffbc850938867", "generation_seconds_batch": 21.03086467203684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3309177160263062, "target_z": 1.8671873591288837, "ce": 5.927301406860352, "genuineness": 4.193545341491699, "identity": 0.5695005059242249, "hypothesis": "Your message arrived after I had left. I I read it twice before putting the phone away. Let me know when we can talk.", "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/emo_Pain/31fedc6248c3c7cb7fecba1d6b9a9fef2ce84d55bcb95459c023484889f83669.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "3204a7f788ed96511817f7c4a4ae7daa56bf3fee219b5af3d4a8c7c78a35564d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "eeeba7a55406127f4f8d097c6164b99558ae0661dd4e3d894ff79f3400f1924c", "generation_seconds_batch": 23.812703671050258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5018618702888489, "target_z": 0.4257812459519794, "ce": 5.974957466125488, "genuineness": 4.7332844734191895, "identity": 0.5747591257095337, "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/emo_Pain/3204a7f788ed96511817f7c4a4ae7daa56bf3fee219b5af3d4a8c7c78a35564d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "330c05bfd1cc55f339da586c3310848d778afeea7eabf822aa614dec0b0ae949", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "6ab916706f09089dff425fa78b929fcb0a1849ab23bf1439df6a3c154c659876", "generation_seconds_batch": 16.47908243793063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 5.95112943649292, "genuineness": 4.469690799713135, "identity": 0.6379990577697754, "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": 0.120000000000001, "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/emo_Pain/330c05bfd1cc55f339da586c3310848d778afeea7eabf822aa614dec0b0ae949.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "3321e555b53bfc09054fdb1d04d1e5c23ae2c4244a5e6a1bda51eafb756bf70a", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "0928ceb320978481c00cf5616b35dd45c62b4027a1e01931664e8e2953537db8", "generation_seconds_batch": 17.143992713070475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06935979425907135, "target_z": -0.326171881881635, "ce": 6.0305562019348145, "genuineness": 4.7081804275512695, "identity": 0.6461688876152039, "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/emo_Pain/3321e555b53bfc09054fdb1d04d1e5c23ae2c4244a5e6a1bda51eafb756bf70a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "3335b5706874409baf20f1c378ff42f4bfb517afe9735c6a4498f15f18b05d61", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e4ef0d6618c1940f9146dc8769549490537fc801d3119093f31b0408aa9663d7", "generation_seconds_batch": 23.526632967987098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11429508030414581, "target_z": -0.2480468721663856, "ce": 5.998785495758057, "genuineness": 4.494794845581055, "identity": 0.4842716157436371, "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/emo_Pain/3335b5706874409baf20f1c378ff42f4bfb517afe9735c6a4498f15f18b05d61.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "334e08d387e5c08d7b7749b483778e100a40ef6060af6f3c1ffcd480984f1b72", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c965945213378fc0355554c1b385b736954812c026e741592b741712421e342b", "generation_seconds_batch": 15.7954231480835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.229123115539551, "genuineness": 4.65797233581543, "identity": 0.6927412748336792, "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/emo_Pain/334e08d387e5c08d7b7749b483778e100a40ef6060af6f3c1ffcd480984f1b72.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "8748be747a90196c47ac99ef447ebf6fc889b36c1769829327e4e62792df163b", "generation_seconds_batch": 16.06137857201975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04464539885520935, "target_z": -0.36914061649915675, "ce": 6.109982967376709, "genuineness": 4.432034492492676, "identity": 0.6621484756469727, "hypothesis": "De weg gaat linksaf na de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant.", "wer": 0.0, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/33f63538ac2c49676cc53462aa2024b47e58cb29b8690774a6580785ecb0e60e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "33fa7054c9f11181d928e71956070d623374ef34bd2ecdd6468154768560cadb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8bff07c44b09d92c427a9024b7c88a02c02fc0b0927661e4af2436baff3737d4", "generation_seconds_batch": 22.793283785926178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.5916601419448853, "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/emo_Pain/33fa7054c9f11181d928e71956070d623374ef34bd2ecdd6468154768560cadb.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "ce62a53c212224629226687e84e5eb0cdf7001eb14bcd5fe302fc314fd53657a", "generation_seconds_batch": 17.73184676002711, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.054755836725234985, "target_z": -0.35156249190395883, "ce": 6.2767791748046875, "genuineness": 3.961331605911255, "identity": 0.29918941855430603, "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/emo_Pain/347b084eb1c87968a3fc14b855043a2de140e81cf0874b7d65525285bfb31414.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "34886c344c207bdbef102529a71aa2c3c846901882a77bfe06461b4878d51bc2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d8d95301face14cffe0cbd19ef1f28c9936782e72324eb4ddb513459a5bde8e6", "generation_seconds_batch": 21.722229861072265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08508715033531189, "target_z": -0.298828118118365, "ce": 6.308549880981445, "genuineness": 4.720732688903809, "identity": 0.4803767204284668, "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/emo_Pain/34886c344c207bdbef102529a71aa2c3c846901882a77bfe06461b4878d51bc2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "34989c1f0a5dae7387957cbdc679c29883ea2d31ddd86714764317c5df78d8b1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "e2a1ffec066b17557fa49ba50404477f492f3af2b9156bcecc2dc799c9fd510f", "generation_seconds_batch": 22.21788643707987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1062413454055786, "target_z": 1.4765624141819635, "ce": 5.700935363769531, "genuineness": 4.15588903427124, "identity": 0.6273939609527588, "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/emo_Pain/34989c1f0a5dae7387957cbdc679c29883ea2d31ddd86714764317c5df78d8b1.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": "34dc2893bb6fc0bbcec751b533c840d01f6a4503d8b0ca52dfd4922e3e73fd7c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0fa763a52ae50c1f0331b6df26000319b3331b47539f98605485d1e2632d6a4d", "generation_seconds_batch": 15.6092861170182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6434080004692078, "target_z": 0.6718749902847506, "ce": 5.4825119972229, "genuineness": 4.413206577301025, "identity": 0.6255325078964233, "hypothesis": "La carretera gira 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/emo_Pain/34dc2893bb6fc0bbcec751b533c840d01f6a4503d8b0ca52dfd4922e3e73fd7c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "35014f5e8d5f533606a8c0e72c1f1fd666f1fd26df62608aff7db5f5b3e5bd92", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5863f782b617900e4f17fe5683cd5dc0a14cbe1bce49018c0811e252a7b1a313", "generation_seconds_batch": 16.812167576979846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8231490850448608, "target_z": 0.9843749255164213, "ce": 5.847874641418457, "genuineness": 4.2249250411987305, "identity": 0.536506712436676, "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/emo_Pain/35014f5e8d5f533606a8c0e72c1f1fd666f1fd26df62608aff7db5f5b3e5bd92.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "35220d4d617da77f30d6d6af453905635a1cba9118c957620e6c3c6779d4a8fc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b30d590a3576a72dfdd33d8081fbb10da09226edbe85f697550863669fdff88d", "generation_seconds_batch": 19.283897335990332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.984916090965271, "target_z": 1.2656249190395883, "ce": 6.109982967376709, "genuineness": 4.833701133728027, "identity": 0.7589380145072937, "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/emo_Pain/35220d4d617da77f30d6d6af453905635a1cba9118c957620e6c3c6779d4a8fc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "3532deee11f7634df57aada638ce4f402f3c84d50783c4dcd576a8737491dae0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "77e1471e3182cd455aea1667ce22768cbbd516096ff76eaf9bda45d901b6e26c", "generation_seconds_batch": 22.669879650929943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 5.903473377227783, "genuineness": 4.632867813110352, "identity": 0.7591567635536194, "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/emo_Pain/3532deee11f7634df57aada638ce4f402f3c84d50783c4dcd576a8737491dae0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "35521f4797ee73d6416a1e96705512513e2cc695dc77574ff20f28528619cae0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2463f587d640282c4d6d50e3790bf498a4fba2aea895a4223fa384d37722e632", "generation_seconds_batch": 21.88280551705975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.009820550680160522, "target_z": -0.4296875016192082, "ce": 5.808161735534668, "genuineness": 4.7081804275512695, "identity": 0.6939790844917297, "hypothesis": "La porte est encore ouverte. Je 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/manual-comparison-v1/audio/emo_Pain/35521f4797ee73d6416a1e96705512513e2cc695dc77574ff20f28528619cae0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "35860ff2af8d5042a2e9fb15ebcae6a51b7be90f46e5f83321e791ef6799812e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "273b8ad0bdf35cec9c59e1dd92d900ea8f614171e1d3e9e2523477d32b3920ad", "generation_seconds_batch": 22.28077194094658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 5.657250881195068, "genuineness": 4.6705241203308105, "identity": 0.7553623914718628, "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/emo_Pain/35860ff2af8d5042a2e9fb15ebcae6a51b7be90f46e5f83321e791ef6799812e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "359d3a3b7f4e7be4e6fc56b8cdf8d47936bfde2372e2354d43fb20d9bf56a182", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2dbd1d4d33aa980539b12056a6aaf4c900ad74609241ab65d8547e5f6fa9a620", "generation_seconds_batch": 23.9398294739658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.46816039085388184, "target_z": 0.3671874627582106, "ce": 5.903473377227783, "genuineness": 4.494794845581055, "identity": 0.5497087240219116, "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/emo_Pain/359d3a3b7f4e7be4e6fc56b8cdf8d47936bfde2372e2354d43fb20d9bf56a182.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "35b3ffdde44450c59c5c1891a083f89acc37df7850ef14669405b25b15eeda27", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f975fad6d73969dd8f5c2fdeb4e538b79a2c8848142229cdffb5940f652aadd8", "generation_seconds_batch": 22.851109010982327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 5.935244083404541, "genuineness": 4.1998209953308105, "identity": 0.6030551195144653, "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/emo_Pain/35b3ffdde44450c59c5c1891a083f89acc37df7850ef14669405b25b15eeda27.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "35d79cc20c83048daf506ae9e411e3463321be9e6bb9a1007bb5ec25612d7bc8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "81decae15f78417f101f00aaeca3d4426bd8b1974b02f388793da75b851cd620", "generation_seconds_batch": 22.850971462088637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1210353672504425, "target_z": -0.23632813107203088, "ce": 5.927301406860352, "genuineness": 4.934117794036865, "identity": 0.7062872648239136, "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/emo_Pain/35d79cc20c83048daf506ae9e411e3463321be9e6bb9a1007bb5ec25612d7bc8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "36389370d2b9a285266d8274977c0123edc8e7720145fd21306d94a45091135e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2846fc4c510b011b9d48d01801fd86449cc3e0fe846eac42173690807fa672bb", "generation_seconds_batch": 23.83019783499185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5690747499465942, "target_z": 2.2812499417085035, "ce": 5.605623245239258, "genuineness": 4.093128204345703, "identity": 0.19851043820381165, "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/emo_Pain/36389370d2b9a285266d8274977c0123edc8e7720145fd21306d94a45091135e.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": "366bf58ceb1373cd980c0f3a87048cc85d19f02366ceeac09e1a2aa2f030d51d", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5afe4464686c3a8ac1720f6de93050ddb9039208011b06bc58d10159b87472d3", "generation_seconds_batch": 17.414270774112083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13507765531539917, "target_z": -0.21191404610551662, "ce": 6.078212261199951, "genuineness": 4.080576419830322, "identity": 0.43534478545188904, "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/emo_Pain/366bf58ceb1373cd980c0f3a87048cc85d19f02366ceeac09e1a2aa2f030d51d.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": "36db90d1094cb90cfdf0546fca350f475819c1b96e8bed07ecaa56747ac7a595", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Llámame cuando tengas un momento.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Escribí el número en el sobre.\n(Natural clear speech.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Natural clear speech.) [3.2 seconds duration] Llámame cuando tengas un momento.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1a9c6b2ce80f314d61e0b17245374b170b0138e125e0c746b439a59f1e2dd072", "generation_seconds_batch": 15.661752338055521, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 5.998785495758057, "genuineness": 4.3881025314331055, "identity": 0.6977758407592773, "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/emo_Pain/36db90d1094cb90cfdf0546fca350f475819c1b96e8bed07ecaa56747ac7a595.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "372967380fd8614d7c6e35f312df0ee8c4765c9fecf1b3a50f10762b7d8c010a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "79ebf798e3f844f04cb189819d5e3285c71f4f0633e43c7c365684d944f1fea6", "generation_seconds_batch": 23.65452500700485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.33990478515625, "target_z": 1.8828123817977989, "ce": 5.982900142669678, "genuineness": 4.281409740447998, "identity": 0.35817283391952515, "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": 0.120000000000001, "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/emo_Pain/372967380fd8614d7c6e35f312df0ee8c4765c9fecf1b3a50f10762b7d8c010a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "37514f76d8faba68ab5950393860f47b952b52b084af6eabea803f4e570eae4f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "db266415969359be45a48ffaf28c63feb718280d4532cdbf5b6ac73c2cc915a5", "generation_seconds_batch": 16.78800477704499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18675321340560913, "target_z": -0.12207032120324426, "ce": 6.205295085906982, "genuineness": 4.494794845581055, "identity": 0.3797153830528259, "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/emo_Pain/37514f76d8faba68ab5950393860f47b952b52b084af6eabea803f4e570eae4f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "37cef3b06aa94c18533730d28a74ecdc8685d67c45be0cd0ad7cf6994a9ca89a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d2be35f5eecbf587e81de8a043e40f7988229fcabf89a0e5f88cbf28c3a58d02", "generation_seconds_batch": 24.02672382502351, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3533854484558105, "target_z": 1.9062500194304988, "ce": 5.967014789581299, "genuineness": 4.300237655639648, "identity": 0.5144497156143188, "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/emo_Pain/37cef3b06aa94c18533730d28a74ecdc8685d67c45be0cd0ad7cf6994a9ca89a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "38151182a92e0fe0d319e3500743f36bd2b0a3d91eedefecf0a08294d880597c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "39246fb8354b75c92e62a21a7c770c86dddad2709361c2788fad21e6608f6c37", "generation_seconds_batch": 16.795798577019013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17439600825309753, "target_z": -0.14355470146567104, "ce": 5.7128496170043945, "genuineness": 4.482243061065674, "identity": 0.5868127346038818, "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/emo_Pain/38151182a92e0fe0d319e3500743f36bd2b0a3d91eedefecf0a08294d880597c.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": "38436a7ed3b23724cca0d05ff44a859213e2d58a7f02c16f4d9e9da6c9a00a22", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8ab603b966a3afdef8c9c36566586398c51d934704988daf8768d24a32ae1d36", "generation_seconds_batch": 23.868913928978145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11429508030414581, "target_z": -0.2480468721663856, "ce": 6.133810997009277, "genuineness": 4.833701133728027, "identity": 0.691520094871521, "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/emo_Pain/38436a7ed3b23724cca0d05ff44a859213e2d58a7f02c16f4d9e9da6c9a00a22.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "38de691958cea0efff91551cc64c1cd3107b0d2c7e24f524660a606287281394", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "438b3ddc064603828c34e2334293c4ac3d162e80952ffb9a25bf47eb28e0db7d", "generation_seconds_batch": 16.31760609499179, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2434840053319931, "target_z": -0.023437508096041174, "ce": 6.348263263702393, "genuineness": 4.65797233581543, "identity": 0.4971959888935089, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/38de691958cea0efff91551cc64c1cd3107b0d2c7e24f524660a606287281394.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "390b3527e60bced6ddbdb2907d0b5bc164e62edbed8a51a46c0d31d4edc6c6b1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a069f4529e864b0129a2fbff7f53ea9b505c9b996b139886b7cab7b7934371a6", "generation_seconds_batch": 22.800526950042695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5411802530288696, "target_z": 0.49414064240648853, "ce": 5.518253803253174, "genuineness": 4.570107460021973, "identity": 0.41680094599723816, "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/emo_Pain/390b3527e60bced6ddbdb2907d0b5bc164e62edbed8a51a46c0d31d4edc6c6b1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "3b4295b775ca77362f8be257cebe8db662baec059b910da01c238b5164d0f09f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "270296dd623a149c980fa59cf4811cc5ca18fb2f7922d81eccaac708cdc0de5d", "generation_seconds_batch": 24.62328832608182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09407420456409454, "target_z": -0.2832031213567815, "ce": 6.141753673553467, "genuineness": 4.833701133728027, "identity": 0.7242531776428223, "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/emo_Pain/3b4295b775ca77362f8be257cebe8db662baec059b910da01c238b5164d0f09f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "3bd2e7a5f2950304e91249842af3d5edcb2488f9d9e77378f6fa865cdd8dcd79", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "791ab4472d688ed2622a5d2f4e8eb1d45210d9e3eabfac4905f75998f5cae8f6", "generation_seconds_batch": 22.596703553106636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8815649747848511, "target_z": 1.0859374692350436, "ce": 5.557967185974121, "genuineness": 4.243753433227539, "identity": 0.5894104242324829, "hypothesis": "La puerta sigue abierta y he dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "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/emo_Pain/3bd2e7a5f2950304e91249842af3d5edcb2488f9d9e77378f6fa865cdd8dcd79.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "14384fe4983eda3ab773eefae17aea1057a526918b155df351541d7922d6a32f", "generation_seconds_batch": 17.1347317219479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.016560837626457214, "target_z": -0.41796876052485354, "ce": 6.284721851348877, "genuineness": 4.469690799713135, "identity": 0.7290947437286377, "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/emo_Pain/3c590f9302048ad206a88d3c2440f1028cc9eea1f1662250baf365be21a6ebfa.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "3c5aef2fdb09f511fe0debcdb65eab70f11390e0f24b60948d571bb30502d2e1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d9cdd23ff6bd07b2f445482f8e749d9957587075b5b7507ce03553b384d219df", "generation_seconds_batch": 22.369334277929738, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41873160004615784, "target_z": 0.28124999352316704, "ce": 5.816103935241699, "genuineness": 4.582659721374512, "identity": 0.5804386138916016, "hypothesis": "La porte. La porte 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/emo_Pain/3c5aef2fdb09f511fe0debcdb65eab70f11390e0f24b60948d571bb30502d2e1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "3d048244d5e72b430a7d70e8783b7483fec80709177b5529a8ad09e9dc87074f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "863115d532c585bccc7fe83c138ff0f0c505f88f43c133143c8ab581644f1193", "generation_seconds_batch": 23.739762646961026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646572351455688, "target_z": 1.5781249579005858, "ce": 5.518253803253174, "genuineness": 4.65797233581543, "identity": 0.6540814638137817, "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/emo_Pain/3d048244d5e72b430a7d70e8783b7483fec80709177b5529a8ad09e9dc87074f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "3d7a25906100b418c36aaa468cf4864d5b761396887a785593bef59f6096b6c9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "65784520e8edb38a40385098df1337cee2c6a5c8b9b53dfeb8c02d1816156049", "generation_seconds_batch": 23.55150776996743, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 5.9431867599487305, "genuineness": 4.457139015197754, "identity": 0.7594844102859497, "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/emo_Pain/3d7a25906100b418c36aaa468cf4864d5b761396887a785593bef59f6096b6c9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "3df7cb2d1a71bef0f66db24f1e2964c3e486b03a2db9a0a158856c1fdd18bec5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c6ff6b53700cf2a3ec18f661b120465e66a94dabda9951f0a4c66f773c507ff0", "generation_seconds_batch": 22.726454457035288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.008697167038917542, "target_z": -0.43164062945282267, "ce": 6.197352409362793, "genuineness": 4.482243061065674, "identity": 0.723691463470459, "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/emo_Pain/3df7cb2d1a71bef0f66db24f1e2964c3e486b03a2db9a0a158856c1fdd18bec5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "3e324840a3bb77c882462217ed7b3ab312ebdd5fac26b41c9e03d9da821fab37", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "58397e6e1370947b66959868f61aacfa05841c7d6dfb4ca2b6bc4e8076e7ff5a", "generation_seconds_batch": 22.587140236981213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5939792394638062, "target_z": 0.5859375728643705, "ce": 5.895530700683594, "genuineness": 4.293961524963379, "identity": 0.30661705136299133, "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/emo_Pain/3e324840a3bb77c882462217ed7b3ab312ebdd5fac26b41c9e03d9da821fab37.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": "3e76952892b4e137b7c65502b54940e6373da29ff21942dac84fc9e8fd34f2e8", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "150c746887fa8189b32db11c83697db0dd05638027d332e5d9b805f6f6f10732", "generation_seconds_batch": 16.938118551974185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055879220366477966, "target_z": -0.34960936407034443, "ce": 6.022613525390625, "genuineness": 4.582659721374512, "identity": 0.6308603286743164, "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/emo_Pain/3e76952892b4e137b7c65502b54940e6373da29ff21942dac84fc9e8fd34f2e8.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": "3e8e62a4fc4d68b65544fc09f6b120c9208041a2da3e55371351fbbe45c26980", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "b5890a6648b9d75073702f7e3db3e4a56716f98b663d05187e9e2f2d940d52f8", "generation_seconds_batch": 17.364065509056672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9624484777450562, "target_z": 1.226562465996627, "ce": 5.816103935241699, "genuineness": 4.7834930419921875, "identity": 0.7317620515823364, "hypothesis": "Seine Nachricht kam, als ich schon weg war. Ich las sie zweimal. Dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/3e8e62a4fc4d68b65544fc09f6b120c9208041a2da3e55371351fbbe45c26980.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "3eab37d716a61030cb93431065b18c20d859fa43c6374487232bc36fb31348c4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6b9858a9ddfef3cf5b1ee79461ebb702deee05d1d26530215e0cef7ea04db7c9", "generation_seconds_batch": 18.148460026015528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15586021542549133, "target_z": -0.17578124595197941, "ce": 6.23706579208374, "genuineness": 4.871357440948486, "identity": 0.6280519962310791, "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/emo_Pain/3eab37d716a61030cb93431065b18c20d859fa43c6374487232bc36fb31348c4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "3eadf5594a5ff72f6be0916894f3730d5000aacc84ca00b690fd80bacfa4d371", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b33286765a959b1c05fa836288037ac5673fb82357ff500192d10d221573a6f8", "generation_seconds_batch": 21.951129827066325, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10081449151039124, "target_z": -0.27148438026242677, "ce": 6.173524379730225, "genuineness": 4.758388519287109, "identity": 0.7101355791091919, "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/emo_Pain/3eadf5594a5ff72f6be0916894f3730d5000aacc84ca00b690fd80bacfa4d371.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "3f0a697946865799b3d7f000daeb53023661cb4c16344397692f98ac44466775", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fcf4b41b2f0f1e7a2bcdd3f56611ae0a59d0a4bc3ab10a51d77f074ab4866c42", "generation_seconds_batch": 22.888168504927307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2656708061695099, "target_z": 0.015136721280012866, "ce": 5.4666266441345215, "genuineness": 4.482243061065674, "identity": 0.6111732721328735, "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/emo_Pain/3f0a697946865799b3d7f000daeb53023661cb4c16344397692f98ac44466775.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "3f0ea2f4aa9e310ca66903dfd71bd2e96f9c1229bd02e3aaca025a7fa3df03c4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0a56a68fa0953729ecaf857c4fd40e01648c244f617b7a4182fbe91bb29ea28d", "generation_seconds_batch": 21.96024641301483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06261950731277466, "target_z": -0.3378906229759897, "ce": 5.982900142669678, "genuineness": 4.570107460021973, "identity": 0.8200170993804932, "hypothesis": "J'ai trouvé la petite euh 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/emo_Pain/3f0ea2f4aa9e310ca66903dfd71bd2e96f9c1229bd02e3aaca025a7fa3df03c4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "3f4ff04ecf6a36b7f2a09af0813da884cc9e551c0ca246e154eb6bef073dbdb4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "578179b19249f74266f1958e7c67960cd1d2cfc6c666a81ec0f3db207ef1deeb", "generation_seconds_batch": 25.72819890396204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24250105023384094, "target_z": -0.025146485235204373, "ce": 5.855817794799805, "genuineness": 4.7458367347717285, "identity": 0.6479845643043518, "hypothesis": "La porte était encore ouverte, 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/manual-comparison-v1/audio/emo_Pain/3f4ff04ecf6a36b7f2a09af0813da884cc9e551c0ca246e154eb6bef073dbdb4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "3f87a1ef57e68ec4e243ce63129c8428d0a8c2c4ee66652f9de967339a5fc119", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c8db2a59fbdfbe241fc308e501494fbfb5d6aa90a90ce3b4b724cecd38400773", "generation_seconds_batch": 22.142778928973712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 6.252951145172119, "genuineness": 4.632867813110352, "identity": 0.48034754395484924, "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/emo_Pain/3f87a1ef57e68ec4e243ce63129c8428d0a8c2c4ee66652f9de967339a5fc119.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": "3ffff7b740ba30c525ccac837868b048f1ed92e0ac0e828bc96efddabc54f685", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Sag mir, wann wir reden können.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Natural clear speech.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wann wir reden können.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5560e1d7ce75d531a24f305fefb6d9ab6319576b4ba49e4f5bb55d69809ac71c", "generation_seconds_batch": 17.404119281913154, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3361630439758301, "target_z": 0.13769534387215954, "ce": 6.205295085906982, "genuineness": 4.821148872375488, "identity": 0.8092026114463806, "hypothesis": "Eine Nachricht kam, als ich schon weg war. Ich las sie zweimal, dann legte ich das Handy weg. Sag mir, wann wir reden können.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/3ffff7b740ba30c525ccac837868b048f1ed92e0ac0e828bc96efddabc54f685.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "404570bdfbefd9adafb2a6a4f79ac817d6cb2609e446e94ed9b84c0f0059b2c3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "47e40a06a8ac0900126f9cfc24be651df8d41ef6ddbb96852e675b5319e676ef", "generation_seconds_batch": 18.326504791039042, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.274938702583313, "target_z": 0.031249993523167062, "ce": 5.895530700683594, "genuineness": 4.720732688903809, "identity": 0.5219032168388367, "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/emo_Pain/404570bdfbefd9adafb2a6a4f79ac817d6cb2609e446e94ed9b84c0f0059b2c3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "4060a07a9d44eb4d1b5a9996cf675b1e343fa1198c35e977d9a9a378dd1e85c6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b517ebe06d5d2f6c711b26b9383ad2abd4cbc6a3ba34bc6f85cf39def375a118", "generation_seconds_batch": 23.65698999899905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31425708532333374, "target_z": 0.09960938997767617, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.3554239869117737, "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/emo_Pain/4060a07a9d44eb4d1b5a9996cf675b1e343fa1198c35e977d9a9a378dd1e85c6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "409837b5fd64741e6b3b24da2c755301427b94bec7bfc1c56ccf45a16e440ec1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a063a3b8d4e5a6f42338d39a40fd2099468ea0bd1c5a42db0277d1f6c12985a2", "generation_seconds_batch": 17.014094367041253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.629927396774292, "target_z": 0.6484374562813776, "ce": 5.808161735534668, "genuineness": 4.570107460021973, "identity": 0.24829936027526855, "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/emo_Pain/409837b5fd64741e6b3b24da2c755301427b94bec7bfc1c56ccf45a16e440ec1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "41d61aafbc57d7fc731ad16286501589e31c4e1f2ff5b33bee090b140cece3a1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b854def4eac34fd45c1cbdde4e2969c0aabb6536f3924878497f968c7a7dd5c0", "generation_seconds_batch": 17.58817168790847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1828213781118393, "target_z": -0.12890625566722883, "ce": 5.982900142669678, "genuineness": 4.243753433227539, "identity": 0.3232199251651764, "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/emo_Pain/41d61aafbc57d7fc731ad16286501589e31c4e1f2ff5b33bee090b140cece3a1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "421074e3dff242a991e14e85ac7ea7c9049b97178130117febb7c9fba4425543", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f50b66b32c49f6dd9cd83b927d31fdf6c4fe74477797e0f3f6579bc917818010", "generation_seconds_batch": 20.67807626002468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 6.173524379730225, "genuineness": 4.758388519287109, "identity": 0.3463226556777954, "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/emo_Pain/421074e3dff242a991e14e85ac7ea7c9049b97178130117febb7c9fba4425543.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "423432ba2489dc463fa8683d754925656dd3319c682c8728801cf8be4a0b1065", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "84b1aaf84b8edb61e95fea8ee403ec1cfdf4c0cf8079b10f6dc6479389c719e7", "generation_seconds_batch": 25.245366242947057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42884203791618347, "target_z": 0.298828118118365, "ce": 5.502368927001953, "genuineness": 4.2249250411987305, "identity": 0.5251208543777466, "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/emo_Pain/423432ba2489dc463fa8683d754925656dd3319c682c8728801cf8be4a0b1065.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": "42560cd3e44576f93ea5b559bc186be23cdcdc68300a842847e88d25204b4439", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "653ef3011a155c44959ecb594d34b6406c6e560867c337ec021028f929e1fe55", "generation_seconds_batch": 15.233560685068369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11766523122787476, "target_z": -0.24218748866554235, "ce": 5.911416053771973, "genuineness": 4.174716949462891, "identity": 0.7009347677230835, "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/emo_Pain/42560cd3e44576f93ea5b559bc186be23cdcdc68300a842847e88d25204b4439.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "426c80383c6d57791e1850dde17c9e396da9e9790732fb22f622ea00a39b0d74", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "86176d4cbc4072f7e01c8bc49f9e1d9a4c6bd3c8e9f2aacf007131c6727e6875", "generation_seconds_batch": 21.367333467002027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9624484777450562, "target_z": 1.226562465996627, "ce": 5.669164657592773, "genuineness": 4.319066047668457, "identity": 0.31662681698799133, "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/emo_Pain/426c80383c6d57791e1850dde17c9e396da9e9790732fb22f622ea00a39b0d74.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "42ce69a5048a7990ad96e2828658732471d67979402cb1d3b8baad308cc717aa", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "67c7a6a765fa4d422eb7587ff2959c5d1be36cd442a41b0d36e2d2fe7833c767", "generation_seconds_batch": 23.59774862404447, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2545278072357178, "target_z": 1.7343749773310848, "ce": 5.6334228515625, "genuineness": 4.356721878051758, "identity": 0.6948171854019165, "hypothesis": "Mais 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/emo_Pain/42ce69a5048a7990ad96e2828658732471d67979402cb1d3b8baad308cc717aa.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "17d505a418a248d2f75afcf4483b36ee7798413cf4d61672767e85d54d6377d5", "generation_seconds_batch": 16.06137857201975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.141753673553467, "genuineness": 4.469690799713135, "identity": 0.6234979033470154, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder tot aan het kleine houten hek. Ik wacht aan de andere kant. Ik.", "wer": 0.09523809523809523, "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/emo_Pain/434f953004ee3838b31e4f505ba0cb9202b34ab37936979fff50edb64a6a996f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "43d209a5f3315ad604098d11d2dea411e88ae99d4b303e512e208e986a05577c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2f664530a4fa06f4065f375b8ae4f9bfbedc1a2ee47c5d22fdabc5df16d17fe4", "generation_seconds_batch": 24.347871454898268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4792041778564453, "target_z": 2.124999922278005, "ce": 5.450741291046143, "genuineness": 4.105680465698242, "identity": 0.5689938068389893, "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/emo_Pain/43d209a5f3315ad604098d11d2dea411e88ae99d4b303e512e208e986a05577c.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "31f0b744f74c1a64bc4ea78a98801b9efeed1fb701072166f941151827d274e5", "generation_seconds_batch": 17.45467336208094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 6.181467056274414, "genuineness": 4.7332844734191895, "identity": 0.6199485659599304, "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/emo_Pain/43d771a5c055e2aea1186dc99f4cdef2b4c6bc8c6b2779ad35a29ca11c75a91c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "43f82d65c35dd8b40bf849fdb34f8510d28dba736fa939ba1a98149908497f42", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d79f9ca263efdae19894beaa0c2d036e786fc0a45f17a89983054567364c99bf", "generation_seconds_batch": 20.978658786974847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.117925643920898, "genuineness": 4.494794845581055, "identity": 0.7926031351089478, "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/emo_Pain/43f82d65c35dd8b40bf849fdb34f8510d28dba736fa939ba1a98149908497f42.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "4407840dce16393075c3d3d097932f8398ebccf8918b9cd28235d34892e76b7e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c24931e09aa251ae9ca38e72688bde2a9bff2864cf39a73d9d9d85a51a1f6133", "generation_seconds_batch": 23.373129783081822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0163708925247192, "target_z": 1.3203126020101188, "ce": 5.855817794799805, "genuineness": 4.645420074462891, "identity": 0.6609898209571838, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. Je quasse.", "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/emo_Pain/4407840dce16393075c3d3d097932f8398ebccf8918b9cd28235d34892e76b7e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "44e1d3aa801254a2cc50b2b8d91ec7152b438d28d0143ac0da21399bc73b13eb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8076ca48ec10de75ef94837dd72de12cf08f63d10787e6e1c198c5571ce1829c", "generation_seconds_batch": 23.908837343915366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5535374879837036, "target_z": 0.5156250744835787, "ce": 6.24500846862793, "genuineness": 4.293961524963379, "identity": 0.5879082679748535, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/44e1d3aa801254a2cc50b2b8d91ec7152b438d28d0143ac0da21399bc73b13eb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "45758c5780ad28b108300b066532633e38778f213e6f4442ac06a1da12b84189", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "04fdf11c0af412c4ecd7693392371b06b19546d4ad98f300cc9266c0e03dd94f", "generation_seconds_batch": 23.265009890892543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4209783673286438, "target_z": 0.2851562491903959, "ce": 5.95112943649292, "genuineness": 4.469690799713135, "identity": 0.49433308839797974, "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/emo_Pain/45758c5780ad28b108300b066532633e38778f213e6f4442ac06a1da12b84189.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "45efc604edda0fb3818557f40f12f0f14f7f364241db05da58c054571cc1ad55", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "09c54f9cda1ce2368a9780feaa7678accf2a8442d9f5b43f35cac57daaf162f9", "generation_seconds_batch": 23.98794989695307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6591353416442871, "target_z": 0.6992187281406889, "ce": 5.283945083618164, "genuineness": 4.432034492492676, "identity": 0.6034122705459595, "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/emo_Pain/45efc604edda0fb3818557f40f12f0f14f7f364241db05da58c054571cc1ad55.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "45f96541ede2f25ea8acea1794feb52e72d71ac2b25c987ae87a0a06cd73c97f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8dbad72ecb249947318fcb26b9333b351c19e484b26bb06a079d114e36600584", "generation_seconds_batch": 16.113576994044706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6389144659042358, "target_z": 0.6640624789502929, "ce": 5.645336627960205, "genuineness": 4.595211505889893, "identity": 0.5585705041885376, "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/emo_Pain/45f96541ede2f25ea8acea1794feb52e72d71ac2b25c987ae87a0a06cd73c97f.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "a47e9f00f10d8075ba88c7070c32cded2d4db19ced94c74c1db1c4c32ee54172", "generation_seconds_batch": 17.23115508200135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.340320587158203, "genuineness": 4.055472373962402, "identity": 0.575996458530426, "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/emo_Pain/45f9ec8497c8691f9e18d794cf98f47c522bfab041b36df5accabf5c5c456412.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "4619b5afbc81134a24d2769ef7db4bd996e229efb05f6eda4484f24e9b4381ab", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e1f176aedc1bac5aac73886c9482e1b00d621081b0c4d5b06e9f4df279b15cb2", "generation_seconds_batch": 22.587140236981213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6411612033843994, "target_z": 0.6679686828028583, "ce": 5.816103935241699, "genuineness": 4.080576419830322, "identity": 0.43140342831611633, "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/emo_Pain/4619b5afbc81134a24d2769ef7db4bd996e229efb05f6eda4484f24e9b4381ab.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "462139639e6e67e429ad236a58bf10997591be208594ee5e6cbdeb53f2058170", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4396ba5d5af72b4f01d125657e3b0de6e0158e2e438846d548fa5636803c5e6a", "generation_seconds_batch": 22.429160495987162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02442452311515808, "target_z": -0.40429686568955264, "ce": 6.022613525390625, "genuineness": 4.331617832183838, "identity": 0.4092431962490082, "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/emo_Pain/462139639e6e67e429ad236a58bf10997591be208594ee5e6cbdeb53f2058170.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "46395512dd0225b6161d046584b728ea183d29c9e413eb732eb8e2cbe3413999", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "73c478e83a926e3eb20dcea21dbc4440dac6cceeaf1414bfa6a74e91acddee67", "generation_seconds_batch": 22.76200983207673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07272994518280029, "target_z": -0.3203124983807918, "ce": 6.332377910614014, "genuineness": 4.6956281661987305, "identity": 0.503618597984314, "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/emo_Pain/46395512dd0225b6161d046584b728ea183d29c9e413eb732eb8e2cbe3413999.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "46760880b2b1d4903b9e5e82675e89f8d71f16ab049252504588211940264544", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "783ef89669fa40221283c121ca6ff17e160d1cd36925d2da0fbdc75aef9d5213", "generation_seconds_batch": 22.43720601592213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1107349395751953, "target_z": 1.4843750291457483, "ce": 6.268836498260498, "genuineness": 4.369274139404297, "identity": 0.5303468108177185, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/46760880b2b1d4903b9e5e82675e89f8d71f16ab049252504588211940264544.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": "468d337a2954feb1cfc794d018c0875d748963726415a7bb0fd173ffac38762d", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "631d317d9534bdcbdb78f3bce094c02a021def77a1f5d954876439d4bf39b3e7", "generation_seconds_batch": 15.233560685068369, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14125624299049377, "target_z": -0.20117188188163498, "ce": 6.173524379730225, "genuineness": 4.444586753845215, "identity": 0.37010055780410767, "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/emo_Pain/468d337a2954feb1cfc794d018c0875d748963726415a7bb0fd173ffac38762d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "46e079bb8acccec5ba0caddce5a137345c34765d120d1cafba4fb5baca7b4e54", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7aeb8231deec4a30ba2593319305570b5e66051997ed304a568a813e2f8c8c6a", "generation_seconds_batch": 22.67614739399869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1982678771018982, "target_z": -0.10205079329286125, "ce": 6.0146708488464355, "genuineness": 4.720732688903809, "identity": 0.68230140209198, "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/emo_Pain/46e079bb8acccec5ba0caddce5a137345c34765d120d1cafba4fb5baca7b4e54.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "47478e27f4ffce8d89b2ac118dc4ece40baae10d2f6197b17b2daa1256645987", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "52412a79679d9ecf0ba66c94f1aaf14b5bcb7028a658f599bfebb890e01718ae", "generation_seconds_batch": 22.44372973893769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6613820791244507, "target_z": 0.7031249319932541, "ce": 5.895530700683594, "genuineness": 4.319066047668457, "identity": 0.020125366747379303, "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/emo_Pain/47478e27f4ffce8d89b2ac118dc4ece40baae10d2f6197b17b2daa1256645987.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "8d57da0c9bf5f3382676f1b395cad684152a2f3be4c54770ba081dc879d29b4f", "generation_seconds_batch": 16.218586139031686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.038498878479004, "genuineness": 4.507347106933594, "identity": 0.7515003681182861, "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/emo_Pain/47d00b404f769e82194cfb39f2be7a6bd3cc5bfaac4f38356b0fada2abdb9f4d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "47d0e016d4c4609262f1f4cd5b66d57113caf8ae9579c578188855d93e3267c8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3488b5bdb5d7c3af46cd030d621794368536303ca692dcf709798c5ca7ea2a73", "generation_seconds_batch": 17.657214643084444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04352201521396637, "target_z": -0.3710937443327712, "ce": 5.7684478759765625, "genuineness": 4.331617832183838, "identity": 0.7510865330696106, "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/emo_Pain/47d0e016d4c4609262f1f4cd5b66d57113caf8ae9579c578188855d93e3267c8.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": "47d37be4b128dee5beda609aa53ae97c09bf723dd0bf632bd49d77ae3f81c15f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2650be1259cabaf0199affa723e54673802e3d8c3df395cf535ab389fdb61c4a", "generation_seconds_batch": 22.69370171998162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12609058618545532, "target_z": -0.22753906877443192, "ce": 5.879645347595215, "genuineness": 4.645420074462891, "identity": 0.6965312957763672, "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/emo_Pain/47d37be4b128dee5beda609aa53ae97c09bf723dd0bf632bd49d77ae3f81c15f.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": "47e86334326e986c55cea46d49119dc568b3cbd59736c727069e475b6282ba01", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8803cb78465989ca505b98b6ca214d490ac1e83ab989d7bdb230af8ce96e82c0", "generation_seconds_batch": 22.361075052060187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.416294813156128, "target_z": 2.015624970854252, "ce": 6.078212261199951, "genuineness": 4.7458367347717285, "identity": 0.6822992563247681, "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": 0.03999999999999915, "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/emo_Pain/47e86334326e986c55cea46d49119dc568b3cbd59736c727069e475b6282ba01.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "47e8b35ab8c64091c6a95c684964388c044defc4313ee0fb8b45a2cf204c3452", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "4e03ce3f38b58b3d89e5f37584aebc1cf50ae5a040027500e0da60a386cc3fc3", "generation_seconds_batch": 22.43536167300772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5241395235061646, "target_z": 2.2031250356225813, "ce": 5.927301406860352, "genuineness": 4.444586753845215, "identity": 0.5972500443458557, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/47e8b35ab8c64091c6a95c684964388c044defc4313ee0fb8b45a2cf204c3452.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "4805ad68a0efd0b5f2c55f8fe27c2e8ead8568d3a43e3b7405b478631c404705", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e025345de40f57b50e3fd6d7d88957b74ed063e15458c5e9725f3b6e40de2c81", "generation_seconds_batch": 23.086049281060696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4030042588710785, "target_z": 0.2539062556672288, "ce": 5.4388275146484375, "genuineness": 4.6203155517578125, "identity": 0.6224846839904785, "hypothesis": "La portée La portée est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt.", "wer": 0.16666666666666666, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/4805ad68a0efd0b5f2c55f8fe27c2e8ead8568d3a43e3b7405b478631c404705.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "48120625d8e89bda3ca9243bbe5730d050a7f4bdd96334320fb6aa065a9c7d10", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "15aaa1f2bfc0b513c4a806d075c4601f8bd78845ffceb8a472dc722b24fba22a", "generation_seconds_batch": 22.503070921986364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 6.268836498260498, "genuineness": 4.469690799713135, "identity": 0.5257764458656311, "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/emo_Pain/48120625d8e89bda3ca9243bbe5730d050a7f4bdd96334320fb6aa065a9c7d10.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "486bdd6506274e41ef66bc54f5f675b6c50561bc5496ae4580dc59d48f3dc78b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "20f120ae57f9b2d902783971bff783ff64d2bf9ed30a41bc5696cf953e294319", "generation_seconds_batch": 25.48191570106428, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5625244975090027, "target_z": 0.531249993523167, "ce": 5.661221981048584, "genuineness": 4.15588903427124, "identity": 0.0786232203245163, "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/emo_Pain/486bdd6506274e41ef66bc54f5f675b6c50561bc5496ae4580dc59d48f3dc78b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "48c56ad489ed91abfe860693e8aec647c8250a26afaeac253631706a70fc75a2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "74879c004cdd2bca0931b068b7f810db9bfefa8834b0cb78e09b7f90153a3898", "generation_seconds_batch": 22.82334410806652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1878765970468521, "target_z": -0.12011719336962985, "ce": 6.109982967376709, "genuineness": 4.7834930419921875, "identity": 0.7698439359664917, "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/emo_Pain/48c56ad489ed91abfe860693e8aec647c8250a26afaeac253631706a70fc75a2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "48df2f55a7f7f23f6f2808a8b1f34db2802720c787bae69e0cf8befa6a1b080d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "43e43b304ea984d0143cf94ce77c6919442c3cd9c1d460dc13512a7a12fbaf90", "generation_seconds_batch": 17.852927301893942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2502242922782898, "target_z": -0.011718767001686463, "ce": 5.510311126708984, "genuineness": 4.595211505889893, "identity": 0.5443371534347534, "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/emo_Pain/48df2f55a7f7f23f6f2808a8b1f34db2802720c787bae69e0cf8befa6a1b080d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "496b4d0329494d4ecbf62d8d1286e5ed9e2e30289db57155cf568afe870a23d4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2130c859a16e85241f1760c303a52c35e3767bb7e8674259841539c94cd35752", "generation_seconds_batch": 30.14840353198815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9354872703552246, "target_z": 1.1796873979898812, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.23885691165924072, "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/emo_Pain/496b4d0329494d4ecbf62d8d1286e5ed9e2e30289db57155cf568afe870a23d4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "49b94a9b954c46f7158db800da9096526b4ea49d788e4a1f0ba56001ac21346c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b6f2bc5811277a5584bf6c92c02f01ec764f75d8d74cee687821955530c56055", "generation_seconds_batch": 23.586297124042176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6229970455169678, "target_z": 2.3749998704633413, "ce": 5.728734970092773, "genuineness": 4.243753433227539, "identity": 0.6789119839668274, "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/emo_Pain/49b94a9b954c46f7158db800da9096526b4ea49d788e4a1f0ba56001ac21346c.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c25f159f46204af7626735f4218b6f6f372323ac0333b77459a19879f286e048", "generation_seconds_batch": 16.815602077986114, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05026230216026306, "target_z": -0.3593750032384165, "ce": 6.141753673553467, "genuineness": 4.18099308013916, "identity": 0.7277764081954956, "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": 0.03999999999999915, "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/emo_Pain/49d468ae6540cc7c1e5b0e6883c25e2b29599458c9de0a5398000eaca62677ab.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "49e89fe8b5794249596e554c2a1bf8267def1ff85a27a293a0c9574cbe88a5a5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "79ff6bed76accd015e3380bd5824a2542ac0a7b0c0fd8797525a227ebbfd705e", "generation_seconds_batch": 18.830093102995306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.578061819076538, "target_z": 2.2968749643774187, "ce": 5.7247633934021, "genuineness": 4.532451152801514, "identity": 0.6346250772476196, "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/emo_Pain/49e89fe8b5794249596e554c2a1bf8267def1ff85a27a293a0c9574cbe88a5a5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "4a9c4000bf7efd299b51cd6490f67277a0462f353d28fce6ec4d51ddce3fa111", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fbc527c503b12dcf87c45a41454f47aa1408c61eedcedbdde11351d8d1271f58", "generation_seconds_batch": 24.021651820046827, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33223119378089905, "target_z": 0.13085938350084322, "ce": 6.38003396987915, "genuineness": 4.494794845581055, "identity": 0.5634379982948303, "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/emo_Pain/4a9c4000bf7efd299b51cd6490f67277a0462f353d28fce6ec4d51ddce3fa111.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "4aa38d597d066a7c97a6606ab6d12225e76e980dc88e71ca1f95bb28968a877a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "af2ca710699a0456f1da3ea1ff4dbc91e007226ccd0fe16ea1ef8bb1445b646b", "generation_seconds_batch": 23.037476580007933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646572351455688, "target_z": 1.5781249579005858, "ce": 5.919358730316162, "genuineness": 4.770940780639648, "identity": 0.7147952318191528, "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/emo_Pain/4aa38d597d066a7c97a6606ab6d12225e76e980dc88e71ca1f95bb28968a877a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "4ad814482423a5026ca394e2a54014c6da8e1af17bff3a1475be50c77bacece6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ba07b69dfd46121d0ea25656d48d83068775dfd04a66f6646bd7da2f719cb7c7", "generation_seconds_batch": 22.540301393019035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23028427362442017, "target_z": -0.04638671480317993, "ce": 5.9431867599487305, "genuineness": 4.607763767242432, "identity": 0.66080641746521, "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/emo_Pain/4ad814482423a5026ca394e2a54014c6da8e1af17bff3a1475be50c77bacece6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "4adcba005580812072f3ccc80fb1a8fba9d15bcb62c63d6e81dec4d9b95e5d15", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ef241cfe465a3c52f386962b3a74cf133773b28c7aa6e6b2d1712994ca13a844", "generation_seconds_batch": 17.430287464987487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19798703491687775, "target_z": -0.1025390687744319, "ce": 6.086154937744141, "genuineness": 4.821148872375488, "identity": 0.7081204652786255, "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/emo_Pain/4adcba005580812072f3ccc80fb1a8fba9d15bcb62c63d6e81dec4d9b95e5d15.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "4b81444f6a8260c65ee572b5c250d714f5378f86bacb6b86ebd647fa3dca6725", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "115e8cfc7eca7efc9cf0f6314e25fb432c0328e1bcbc39b8ea68989cc66834a6", "generation_seconds_batch": 22.32786624704022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3356013298034668, "target_z": 0.13671874109435472, "ce": 6.109982967376709, "genuineness": 4.808597087860107, "identity": 0.5695229768753052, "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/emo_Pain/4b81444f6a8260c65ee572b5c250d714f5378f86bacb6b86ebd647fa3dca6725.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "4b89152f4a07446fd17088794718459b86369159a2b39c7810b8a96b50e10a9d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9492667d81702343a1d4ad5391367714043c75063205928e4c3ea367d82e3bf9", "generation_seconds_batch": 22.87817033799365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7018238306045532, "target_z": 0.7734374303740459, "ce": 5.7684478759765625, "genuineness": 4.093128204345703, "identity": 0.43327808380126953, "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/emo_Pain/4b89152f4a07446fd17088794718459b86369159a2b39c7810b8a96b50e10a9d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "4bcbe73c52a1affad832547ab90a9a693c9c2499454520efafd6479697aa196f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "1cd962bd5a10f999dd75ab7d1e6632e44af66cf2ae2680baaa330d1c3b1c21d7", "generation_seconds_batch": 22.554827383020893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.735525369644165, "target_z": 0.8320313171971417, "ce": 5.744620323181152, "genuineness": 4.319066047668457, "identity": 0.5990424752235413, "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/emo_Pain/4bcbe73c52a1affad832547ab90a9a693c9c2499454520efafd6479697aa196f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "4bdf72365095a1b3af6326aa74852d637aa88121fe767532f8cf52e754c0670f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "09beb8cc5298e499d4af962555751a5f80314ce0775deeb387a19fde4bcc446f", "generation_seconds_batch": 20.978658786974847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09407420456409454, "target_z": -0.2832031213567815, "ce": 6.157639026641846, "genuineness": 4.68307638168335, "identity": 0.6792819499969482, "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/emo_Pain/4bdf72365095a1b3af6326aa74852d637aa88121fe767532f8cf52e754c0670f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "4c5ac913ffa8e4ac90497707206f7d74c8c27d0ab0eb8a99e7e8cff2c3486676", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2928db53d63ae888e29c3672468f5de77ab13b548044a8e572f666486892a606", "generation_seconds_batch": 24.208149040932767, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.35076698660850525, "target_z": 0.16308592798715163, "ce": 5.506340026855469, "genuineness": 4.419482707977295, "identity": 0.5968390703201294, "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/emo_Pain/4c5ac913ffa8e4ac90497707206f7d74c8c27d0ab0eb8a99e7e8cff2c3486676.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": "4d5f6bf48eb43025aeeef5635eb4a399581c34a5055a25bb1275f01d515b024b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7367e670e1629a6b119572c32bf3aad62d4d7f0d5422bf3c5458d58bbd3cc859", "generation_seconds_batch": 17.972136022988707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 5.621508598327637, "genuineness": 4.432034492492676, "identity": 0.7001758813858032, "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/emo_Pain/4d5f6bf48eb43025aeeef5635eb4a399581c34a5055a25bb1275f01d515b024b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "4d69a6c185d8db2d3054657efbf49469ec543bb35a47c7a8f1031ab4b3faa002", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7c1386e3506a2094984e493c6b1bd9e0999a85e63271473f329a4d6661ed18f5", "generation_seconds_batch": 24.373335314099677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.053632453083992004, "target_z": -0.35351561973757323, "ce": 5.998785495758057, "genuineness": 4.444586753845215, "identity": 0.6761436462402344, "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/emo_Pain/4d69a6c185d8db2d3054657efbf49469ec543bb35a47c7a8f1031ab4b3faa002.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "4dd948940a3b2de4ad3d28262eb3975262966536489e97a987e7e7da1c1647e6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0467085ebdf7f7eec2fa2334ba22ae52f67be4ff1419d4c95c86cc5255dd296d", "generation_seconds_batch": 24.41671281994786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19124674797058105, "target_z": -0.11425780986878661, "ce": 5.919358730316162, "genuineness": 4.7081804275512695, "identity": 0.5250747203826904, "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/emo_Pain/4dd948940a3b2de4ad3d28262eb3975262966536489e97a987e7e7da1c1647e6.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": "4e2e760e56f6e380ffa3df2f061a5100d67d1d2954e96a51be22f5c05d13a773", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "34d9656cb17f33089f6211519242bc4874bd75bf0ac8f638b180c789bf5860ed", "generation_seconds_batch": 17.221479989006184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1148567646741867, "target_z": -0.24707032120324426, "ce": 5.919358730316162, "genuineness": 4.645420074462891, "identity": 0.7616997957229614, "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/emo_Pain/4e2e760e56f6e380ffa3df2f061a5100d67d1d2954e96a51be22f5c05d13a773.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "4e7e2e96b5e9c283cbb45d7e10f4bd7eda16c746eca96af62dab14e4164132da", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2ab876aab57e95727665a1c3747770392d9ab507f4ca99e419ec7f89186519c3", "generation_seconds_batch": 22.548240453004837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29628297686576843, "target_z": 0.06835939645450911, "ce": 6.109982967376709, "genuineness": 4.268857479095459, "identity": 0.6109192371368408, "hypothesis": "He encontrado la cajita. Estaba junto, 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/emo_Pain/4e7e2e96b5e9c283cbb45d7e10f4bd7eda16c746eca96af62dab14e4164132da.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "4eb6d637145314d0740d83b45fe0ae6cfb53ff4eac51312e77f35108a8cb9be5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6ed9e75136c773d5c7165756fa84b3f76be331ee103aca9a0c9cb0e222c2eea9", "generation_seconds_batch": 24.212196702021174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20416563749313354, "target_z": -0.09179687864321853, "ce": 5.824047088623047, "genuineness": 4.519899368286133, "identity": 0.718184232711792, "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/emo_Pain/4eb6d637145314d0740d83b45fe0ae6cfb53ff4eac51312e77f35108a8cb9be5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "4f6df0e462752c9e0c3c903c7b95066d2eaa4204cbcddc3a4868082f4b8c8395", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6c285c35f0f95d3d99e09eee744820d47c7512e303a7dc6e8c9e35feccbfa1d2", "generation_seconds_batch": 26.808179888990708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07048317790031433, "target_z": -0.3242187540480206, "ce": 5.998785495758057, "genuineness": 4.65797233581543, "identity": 0.7740018963813782, "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/emo_Pain/4f6df0e462752c9e0c3c903c7b95066d2eaa4204cbcddc3a4868082f4b8c8395.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "4f7a0ce79f6877d3bc14680d234dc83f2cd99ed2daa0faf2dcd0b5c66448a6f8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "33119b2cb63f08d5ea8486788a2ed2f5a4081423e2896e7e2703e54a0f3b0434", "generation_seconds_batch": 23.038734287023544, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17102587223052979, "target_z": -0.1494140590591825, "ce": 6.09409761428833, "genuineness": 4.959221839904785, "identity": 0.7834295034408569, "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/emo_Pain/4f7a0ce79f6877d3bc14680d234dc83f2cd99ed2daa0faf2dcd0b5c66448a6f8.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": "4f9718189c4c14e52626f90f76545dd85a735ebf6ed668a416e318672e57e836", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "532d99cf213ee5c54255552df6c440f5bf25fd175588e95b53e1c40f186b4fe4", "generation_seconds_batch": 25.250903810025193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3103252351284027, "target_z": 0.09277342960635986, "ce": 6.09409761428833, "genuineness": 4.425758361816406, "identity": 0.6929435133934021, "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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/4f9718189c4c14e52626f90f76545dd85a735ebf6ed668a416e318672e57e836.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": "4ff5d2d7efb7ca3883de939ba797bee98ea33ebba3852bdcb14d8c8333a4697f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "964f5ca0ec670f191dc97cd19e51f7294376c010f247de655f04777a0e44f7a5", "generation_seconds_batch": 17.588278732961044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2696026563644409, "target_z": 0.021972681651329182, "ce": 6.09409761428833, "genuineness": 4.419482707977295, "identity": 0.3029720187187195, "hypothesis": "The road turns left behind the bridge. Keep going until you r reach the small wooden gate. I will meet you on the other side.", "wer": 0.041666666666666664, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/4ff5d2d7efb7ca3883de939ba797bee98ea33ebba3852bdcb14d8c8333a4697f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "5010b37960b5a6b873d737cc7fa0aa4514193c45b7f2ac6b61579ea1245a2317", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "970355e9ea3d723d74bee2ce814736a6e6a05a3eb403f6adf256a4c3b032c220", "generation_seconds_batch": 22.800526950042695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3844684362411499, "target_z": 0.22167965936625691, "ce": 5.855817794799805, "genuineness": 4.406930446624756, "identity": 0.5444235801696777, "hypothesis": "La puerta sigue abue abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "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/emo_Pain/5010b37960b5a6b873d737cc7fa0aa4514193c45b7f2ac6b61579ea1245a2317.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": "506a25d57f1c1ddb54b19743b6ad1dbe23ba8f563c64e5c145bd783256533d87", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ff45c8cd52cd8281a8fa2e6300fa9b9b879c67162dfc0ad5d0c15c799f09dbff", "generation_seconds_batch": 17.296011709957384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13002242147922516, "target_z": -0.22070313431044736, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.40542787313461304, "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/emo_Pain/506a25d57f1c1ddb54b19743b6ad1dbe23ba8f563c64e5c145bd783256533d87.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": "50722755f8ae442c75941d8726d1668dc2c20dba49b07d890dd6c78fc8eebb4e", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9256509811a1e04496feea277813bfa14eeae26fc5f78cf0864d4a5ff335201", "generation_seconds_batch": 17.879899521009065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 6.117925643920898, "genuineness": 4.457139015197754, "identity": 0.512669026851654, "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/emo_Pain/50722755f8ae442c75941d8726d1668dc2c20dba49b07d890dd6c78fc8eebb4e.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": "51a0493337de411e38e0bbfe49f5f4b92bdb9447f57a8ff56ef0a61c966f61fe", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d75a0b61d4d67073a31c168adc18a9aba542386b6d5941d5a2d15164a787c44f", "generation_seconds_batch": 18.299639372969978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08958066999912262, "target_z": -0.2910156326912391, "ce": 6.221180438995361, "genuineness": 4.394378662109375, "identity": 0.515171468257904, "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/emo_Pain/51a0493337de411e38e0bbfe49f5f4b92bdb9447f57a8ff56ef0a61c966f61fe.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "51a4e2ec49c629a7d9bfd1ffc2642e479893344458639586444fa6ff60fce30d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7323019c5cdf7fb93ae565328ac929ba1d1af1e4d2255812aaa690fc75c2f498", "generation_seconds_batch": 23.013096615090035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9354872703552246, "target_z": 1.1796873979898812, "ce": 5.577824115753174, "genuineness": 4.557555198669434, "identity": 0.47800469398498535, "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/emo_Pain/51a4e2ec49c629a7d9bfd1ffc2642e479893344458639586444fa6ff60fce30d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "51e62774484dae64f39ab8382ce27088e58c9cc49608a71e0d25dc88d23ae9b7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d6f1853cb2369121d8c287d6def869deeb016363e7b92d4dbc330a4a96223de7", "generation_seconds_batch": 24.347871454898268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.065799593925476, "target_z": 1.4062499158011719, "ce": 5.4944257736206055, "genuineness": 4.425758361816406, "identity": 0.5026183724403381, "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/emo_Pain/51e62774484dae64f39ab8382ce27088e58c9cc49608a71e0d25dc88d23ae9b7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "523130b9e14ea3ea59a01fbd1c489e379c0eedc7eb53d27abef967b99f2ca687", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "40a1f4c59a7f9d05198a37e8e7a6940059461571fc54e32814dfc3ca09e760dd", "generation_seconds_batch": 22.09440491499845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24825838208198547, "target_z": -0.015136721280012866, "ce": 5.974957466125488, "genuineness": 4.65797233581543, "identity": 0.4802395701408386, "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/emo_Pain/523130b9e14ea3ea59a01fbd1c489e379c0eedc7eb53d27abef967b99f2ca687.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "52b01308479416755927628d75b9c8ebe94ea801d8569238c60858761aa4ef9a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "faf17787de4e8081cddbebe3e0409e6481f847037760f148e76ba6f5e2d0f82a", "generation_seconds_batch": 17.494559879996814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08284038305282593, "target_z": -0.3027343737855938, "ce": 6.157639026641846, "genuineness": 4.68307638168335, "identity": 0.5808877944946289, "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/emo_Pain/52b01308479416755927628d75b9c8ebe94ea801d8569238c60858761aa4ef9a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "52b43c785cea2c4b9687fbbe02ff77f9f0e89a2a62176fa937721b1cecd4ece7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6acde5908a8bb847726e6b1d513cb5ad63a0dc9d50d2b3d7252a197f441c11ee", "generation_seconds_batch": 24.08801390195731, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.259021282196045, "target_z": 1.7421873850362153, "ce": 5.808161735534668, "genuineness": 4.218649387359619, "identity": 0.49419617652893066, "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/emo_Pain/52b43c785cea2c4b9687fbbe02ff77f9f0e89a2a62176fa937721b1cecd4ece7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "52d4ae4878750455daeb6e0f6778cfd2bfcc9731c9475e3bcb9a11e7fa9709b5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5465a994e755c3470345e828ab80d6034722dcc16e21cc96530b70521e82ad82", "generation_seconds_batch": 23.781702080043033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.42434853315353394, "target_z": 0.29101565859857087, "ce": 5.7406487464904785, "genuineness": 4.758388519287109, "identity": 0.48465994000434875, "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/emo_Pain/52d4ae4878750455daeb6e0f6778cfd2bfcc9731c9475e3bcb9a11e7fa9709b5.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": "5316a183f41e4e8d297217a9acdeac794ff2d30cbf87f90c259ccd2e44fbf266", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4dde32753b36ed2ff3eef698a25c2d71cc6bd4a8d3705fc448c6ea316f346dcc", "generation_seconds_batch": 22.944967439048924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9804226160049438, "target_z": 1.2578125113344576, "ce": 5.752562999725342, "genuineness": 4.055472373962402, "identity": 0.14980542659759521, "hypothesis": "De sleutels liggen op de kuikentafel. Ik keek in de en daarna, naast het raam, neem ze mee als je weggaat.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/5316a183f41e4e8d297217a9acdeac794ff2d30cbf87f90c259ccd2e44fbf266.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "533e773f455b3e7ee9e331b8d62c456a9b7a4370f9a3fa65a91901f2bd7ad5c1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a6ba1cdeae1090fb02ab0a0a4d655eedf954c0ff383a09146f1c908c24cc5c51", "generation_seconds_batch": 23.447729049948975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.334477961063385, "target_z": 0.13476563916807205, "ce": 5.804190158843994, "genuineness": 4.532451152801514, "identity": 0.7641445398330688, "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/emo_Pain/533e773f455b3e7ee9e331b8d62c456a9b7a4370f9a3fa65a91901f2bd7ad5c1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "538b7f4e01ede4a8088a130a677f363640cbed3d181058dc7c202d977d117c0c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bd8582b4a9ed8b75ce3d40330c0f2610954bcdb756559b9d028e428058bbbf68", "generation_seconds_batch": 16.470614130957983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.749005913734436, "target_z": 0.8554687475711876, "ce": 5.95112943649292, "genuineness": 4.425758361816406, "identity": 0.5738465189933777, "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/emo_Pain/538b7f4e01ede4a8088a130a677f363640cbed3d181058dc7c202d977d117c0c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "53d2aaa2fa2d71d46f5abe1921696d56b92d781f75a52b110298439f07b54815", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6ed6580784172ca5183c66e74b8bb0d9931f9f067572738d3b1b11ab857eb273", "generation_seconds_batch": 22.596703553106636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1376961469650269, "target_z": 1.531250097152494, "ce": 5.700935363769531, "genuineness": 4.482243061065674, "identity": 0.6233845949172974, "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/emo_Pain/53d2aaa2fa2d71d46f5abe1921696d56b92d781f75a52b110298439f07b54815.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": "541adce2d2ae8e3b3ae7cf734578c224bab9f93b6caa136c4b55ce820859a2d7", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a4439690fe0702517d5708c9d78586ab4bc107783e55ef90eed260c60ab4d4ff", "generation_seconds_batch": 22.78990407299716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3895236849784851, "target_z": 0.2304687734785194, "ce": 5.839932441711426, "genuineness": 4.2249250411987305, "identity": 0.06410113722085953, "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/emo_Pain/541adce2d2ae8e3b3ae7cf734578c224bab9f93b6caa136c4b55ce820859a2d7.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": "542bf9395bc0d7a5589416b719e3ab682b230d33eeccae26b2ded8eb85950e61", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "efb9f32a9abb8a66230e256d4b3c3eba02a168e636b270724872ce2dd84fed48", "generation_seconds_batch": 17.164312410983257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17158755660057068, "target_z": -0.14843750809604117, "ce": 6.0146708488464355, "genuineness": 4.632867813110352, "identity": 0.741101861000061, "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/emo_Pain/542bf9395bc0d7a5589416b719e3ab682b230d33eeccae26b2ded8eb85950e61.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "9557d1d9ecb91a10f30c63c1c652e9f7e04328aa978efc3f405c40138271f17e", "generation_seconds_batch": 17.23115508200135, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.054755836725234985, "target_z": -0.35156249190395883, "ce": 6.324435234069824, "genuineness": 4.344170093536377, "identity": 0.5549522042274475, "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/emo_Pain/546203f143d2dbd908be3097fd76d41d76e93afd1487fa15c64a3d119dff5d49.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "54aec0ede53d7b59c2abdec482280743fbdf597da55b54b0e991549c9119010b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5db013cfaa163d7e95f85758cc3294e983e3f01a6dfea79fedf7783fd2ecfc90", "generation_seconds_batch": 22.82334410806652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38783860206604004, "target_z": 0.22753906877443192, "ce": 5.927301406860352, "genuineness": 4.6705241203308105, "identity": 0.5748957991600037, "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/emo_Pain/54aec0ede53d7b59c2abdec482280743fbdf597da55b54b0e991549c9119010b.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": "5521b80c29060a7c3c0922887b5685f111fa352f3b77baa7de16aeda6bfe6927", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e8bc5b6322512a92e0292c281f7ed560cb30eef67c2fbcae68e794b1dabee1c7", "generation_seconds_batch": 17.2589604689274, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1197220087051392, "target_z": 1.5000000518146634, "ce": 5.784333229064941, "genuineness": 4.444586753845215, "identity": 0.7777958512306213, "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/emo_Pain/5521b80c29060a7c3c0922887b5685f111fa352f3b77baa7de16aeda6bfe6927.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": "55739f4656aa3331915b9188a33bb974b66c74202a70994b579236e48d0eb7f4", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "c5b0fb00ff8f8a838fd7b68adcf22b688876787dba0a75ce76b274bed929ffb3", "generation_seconds_batch": 17.114094709977508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9939031600952148, "target_z": 1.2812499417085035, "ce": 5.863759994506836, "genuineness": 4.086852550506592, "identity": 0.21655821800231934, "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/emo_Pain/55739f4656aa3331915b9188a33bb974b66c74202a70994b579236e48d0eb7f4.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": "558a66238e2d16f19e54763941797c2637154784e39f8ebe3cd02d85d8071998", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dbc6b37a8bed280b17477a9c914cdbb7da7c67cdf0d3d9b394f745625330d29f", "generation_seconds_batch": 22.69370171998162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4580499529838562, "target_z": 0.34960933816301265, "ce": 5.5539960861206055, "genuineness": 4.281409740447998, "identity": 0.6794464588165283, "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/emo_Pain/558a66238e2d16f19e54763941797c2637154784e39f8ebe3cd02d85d8071998.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "55f20bd5cac74ff7bbd4776a39b05f9a91c33d43f061a10b77bbc653ba08bf0c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "61b7e65e37fae2363c5668ba7f819949b96c3dace775d40888a739d63af671ae", "generation_seconds_batch": 17.062629845924675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.252951145172119, "genuineness": 4.595211505889893, "identity": 0.3803732395172119, "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/emo_Pain/55f20bd5cac74ff7bbd4776a39b05f9a91c33d43f061a10b77bbc653ba08bf0c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "55f2903724c06b4b1ead075aa5934e10cbd061b22ad3ed65e101f2b9e8bccfad", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ff1e1045620700b9ae480b9aebc7d8880154256af0a1ecadb587f2bd8fd6fa37", "generation_seconds_batch": 22.930587901035324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3174371719360352, "target_z": 1.8437499287548378, "ce": 5.855817794799805, "genuineness": 4.319066047668457, "identity": 0.10168501734733582, "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/emo_Pain/55f2903724c06b4b1ead075aa5934e10cbd061b22ad3ed65e101f2b9e8bccfad.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "56a468a5eb8b31744bd305c0342a2c73110dd4ffee43f8afc1f2606364370fb7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "37879cc9353ada72efc65ff8786d00d79ec5ddee6387a139edc2ed941a4f7bef", "generation_seconds_batch": 24.103807280072942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.155670166015625, "target_z": 1.5624999352316706, "ce": 5.530168056488037, "genuineness": 4.137060642242432, "identity": 0.6862231492996216, "hypothesis": "J'ai trouvé la petite boîte. J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.2631578947368421, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/56a468a5eb8b31744bd305c0342a2c73110dd4ffee43f8afc1f2606364370fb7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "56ca2413d234030851a7a384e6a8128be099b811ff2fc8e6ae9d47b29b796831", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ed4df7fd59997c1d0e931d2df5f95e15d97199239dbefa78e4f4bc1d895d83dc", "generation_seconds_batch": 22.934741495992057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1575452834367752, "target_z": -0.17285156715522368, "ce": 5.871703147888184, "genuineness": 4.68307638168335, "identity": 0.5273855924606323, "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/emo_Pain/56ca2413d234030851a7a384e6a8128be099b811ff2fc8e6ae9d47b29b796831.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": "56e30c89bbf4ed6a602291f6762e2cc658a2a18914552bfee0a3159c5e9d8508", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] I will meet you on the other side.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The road turns left behind the bridge.\n(Natural clear speech.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Natural clear speech.) [3.2 seconds duration] I will meet you on the other side.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "28142c9ab5e54ce5ca4aaf356b90debdc11af35231a77c355bf2b725f2e15788", "generation_seconds_batch": 17.588278732961044, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11878859996795654, "target_z": -0.2402343867392597, "ce": 6.09409761428833, "genuineness": 4.18726921081543, "identity": 0.5499488115310669, "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/emo_Pain/56e30c89bbf4ed6a602291f6762e2cc658a2a18914552bfee0a3159c5e9d8508.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "57e9d0810d1cd5ff6e4729a46a137be2081d263e38a35d74af9faa4d490ddc32", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "515498c3d504be98bfbe9b82eea2a910063d2cb006c3bbe8c7cc7b17292a41ec", "generation_seconds_batch": 17.690550341969356, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.23706579208374, "genuineness": 4.632867813110352, "identity": 0.7937527298927307, "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/emo_Pain/57e9d0810d1cd5ff6e4729a46a137be2081d263e38a35d74af9faa4d490ddc32.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "585c5b1e9a734f6c8991c9d50d151316826684730fbda120bcf11d2776c4bec0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7a32edb2bed253c053927c135551fa7d5ca7f5e09a5e32a34a35c24ca43ef50f", "generation_seconds_batch": 23.265009890892543, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2894022464752197, "target_z": 0.056396478758371435, "ce": 6.221180438995361, "genuineness": 4.419482707977295, "identity": 0.643924355506897, "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/emo_Pain/585c5b1e9a734f6c8991c9d50d151316826684730fbda120bcf11d2776c4bec0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "58bba2c97258ce26ce66786d90d57a101912480bf38c92fbb47b2b6ea1e0390b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b3cf83ef0831ecbcc40175817fb2f1ee5741dc1ce7fee9152a91a54798b03010", "generation_seconds_batch": 24.212196702021174, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 6.109982967376709, "genuineness": 4.7834930419921875, "identity": 0.7727895975112915, "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/emo_Pain/58bba2c97258ce26ce66786d90d57a101912480bf38c92fbb47b2b6ea1e0390b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "5900ff95d2773a118aed1076e1ee8d8d6288107c56a1eb9dac684a45e556751c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4b17314f8b5d76896fa5ee8447eb06a430c52446d5799d491cedff6466c3741c", "generation_seconds_batch": 18.830093102995306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4612300395965576, "target_z": 2.093749876940174, "ce": 5.605623245239258, "genuineness": 4.268857479095459, "identity": 0.6269490718841553, "hypothesis": "La porte était encore ouverte. 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/manual-comparison-v1/audio/emo_Pain/5900ff95d2773a118aed1076e1ee8d8d6288107c56a1eb9dac684a45e556751c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "5967fa19f7752a1850662f9d5a34a843a84a7796c915ff2e41793a8c00297f61", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "37608e38a8a2d39cf0ad78c0187e0f36833b60b05a398c6a6a3b0f9989628c07", "generation_seconds_batch": 21.364603562047705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07610009610652924, "target_z": -0.31445311487994854, "ce": 6.38003396987915, "genuineness": 4.432034492492676, "identity": 0.5990065336227417, "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/emo_Pain/5967fa19f7752a1850662f9d5a34a843a84a7796c915ff2e41793a8c00297f61.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "596ba172bcc8bca24c77768c08e2420d067ace3360b11dc9067e558439c8cdab", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "98d2c133bec7beb51ee94f67cc97c54bd0cf03a5e79878143efdd464c78f28e3", "generation_seconds_batch": 24.874704266083427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 6.292664527893066, "genuineness": 4.7834930419921875, "identity": 0.6345667243003845, "hypothesis": "He encontrado la cajita. Estaba. estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/596ba172bcc8bca24c77768c08e2420d067ace3360b11dc9067e558439c8cdab.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "59b9aab77faa1c5a6d4bcd7f43dc5365ba588fe972974af4742880d3a6410806", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "29f1121ee8c424936292406e957cf6b15f14727e9b2204ede14424106e05e912", "generation_seconds_batch": 22.329000617028214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.133954256772995, "target_z": -0.2138671998464628, "ce": 5.839932441711426, "genuineness": 4.607763767242432, "identity": 0.5719558000564575, "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/emo_Pain/59b9aab77faa1c5a6d4bcd7f43dc5365ba588fe972974af4742880d3a6410806.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "59f12de94fc480fb64ec0239411ad375d145670e7c0f0a8a1c6b8d88c56953cb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9fa9cb234444c5390cc17f11d49c92ae758fe9845351d1f82f8ae053306b62c5", "generation_seconds_batch": 24.874704266083427, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07272994518280029, "target_z": -0.3203124983807918, "ce": 6.165581703186035, "genuineness": 4.6203155517578125, "identity": 0.7655118107795715, "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/emo_Pain/59f12de94fc480fb64ec0239411ad375d145670e7c0f0a8a1c6b8d88c56953cb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "59fd92c675001dacdac0d8291de7c3070177ac9c218a87ae367307f28562522e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "4c5d5489c318b1c39f29cd5747eea35e02bd9e416909c24f5e7d766caef689f5", "generation_seconds_batch": 23.83019783499185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1466830968856812, "target_z": 1.5468749125627552, "ce": 5.403085231781006, "genuineness": 4.18726921081543, "identity": 0.3565039038658142, "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/emo_Pain/59fd92c675001dacdac0d8291de7c3070177ac9c218a87ae367307f28562522e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "5a895891fc0ea5443fbc2bbf71a0e6a4ef1ec7e6af8367723db1ee11cb12e454", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "57b264778faadff5098e75db25439f4dd3120457d3e7dfbf07a1410adb3ba753", "generation_seconds_batch": 17.014094367041253, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40075749158859253, "target_z": 0.25, "ce": 5.681078910827637, "genuineness": 4.65797233581543, "identity": 0.7201216220855713, "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/emo_Pain/5a895891fc0ea5443fbc2bbf71a0e6a4ef1ec7e6af8367723db1ee11cb12e454.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": "5ae45593c1bffc8e79a76b934e375e77d56cf41501b0bed9d2ade30bd89449d8", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8c24af60ad08684c71a049bd71ed7424c38a5ce60620a51a5aaff9dc0876344a", "generation_seconds_batch": 23.107928665005602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13676270842552185, "target_z": -0.20898439321609263, "ce": 6.213237762451172, "genuineness": 4.419482707977295, "identity": 0.5941609144210815, "hypothesis": "Someone moved a little blue chair. It used to stand here beside th the tall bookshelf. We can put it back after lunch.", "wer": 0.09090909090909091, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/5ae45593c1bffc8e79a76b934e375e77d56cf41501b0bed9d2ade30bd89449d8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "5b5e8c53deace2a2b73a2d3aab9b06a7b7a2b41d1fc89f7b41c60d335bc53fcf", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "fcc12dc93c0ee4fcaceb1858bb0e69de4f21d3a78af4b726548277752fa40c73", "generation_seconds_batch": 24.21055451501161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.23655366897583, "target_z": 1.7031249319932542, "ce": 5.756534099578857, "genuineness": 4.507347106933594, "identity": 0.6013853549957275, "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/emo_Pain/5b5e8c53deace2a2b73a2d3aab9b06a7b7a2b41d1fc89f7b41c60d335bc53fcf.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "5b91ef98b9ee63f320998c850cc7eb5127086fd4b613ecf129f89fef26316296", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "188e1728941ea5e0979bd5d05679532c63d351bbfc7427a490bbc991ea614c5d", "generation_seconds_batch": 21.42293769295793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07497671246528625, "target_z": -0.31640624271356294, "ce": 6.09409761428833, "genuineness": 4.6705241203308105, "identity": 0.6422995328903198, "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/emo_Pain/5b91ef98b9ee63f320998c850cc7eb5127086fd4b613ecf129f89fef26316296.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "5bac6645360b34ccb223a92b76552924be91f2d7b3531a46cb01a4c0ff00a007", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "391bd93a357a856644af5418ab9e439887654ebc73e45dbd7d00f4ed3d38376d", "generation_seconds_batch": 21.330394521937706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8298894166946411, "target_z": 0.9960937443327712, "ce": 6.09409761428833, "genuineness": 4.507347106933594, "identity": 0.27635931968688965, "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/emo_Pain/5bac6645360b34ccb223a92b76552924be91f2d7b3531a46cb01a4c0ff00a007.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "5bfbd4a5a1d2d8e701c82014c3a8483ff67e35521d26ffeacc7318a5f54b5407", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "eaf05a091484af038ad2d500eb426c66161c6ec4347685242ac61380f87d7f92", "generation_seconds_batch": 26.986887359060347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3066742420196533, "target_z": 0.08642577062394596, "ce": 5.804190158843994, "genuineness": 4.68307638168335, "identity": 0.8092355728149414, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la fille fenêtre. Nous pouvons regarder ensemble ce que 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/emo_Pain/5bfbd4a5a1d2d8e701c82014c3a8483ff67e35521d26ffeacc7318a5f54b5407.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "5c187f0972600800838b03a4b199e8fe7e3e99aaa01f0d1a7828a6fafc0a7d04", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8dab2d44b6d4f8fb6581893406d883378d33499b31ed442c78dc87ba79702fa1", "generation_seconds_batch": 21.73654729896225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21764622628688812, "target_z": -0.06835937054717736, "ce": 5.7684478759765625, "genuineness": 4.68307638168335, "identity": 0.5911754369735718, "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/emo_Pain/5c187f0972600800838b03a4b199e8fe7e3e99aaa01f0d1a7828a6fafc0a7d04.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "5c723e8c570bd34e8b8246e103c8f7d38a72933d0f3728f61b89d5fc80cbba1a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "acd4e9008092fc9520573cf100cfeaaa9cace4ef1f9ce0b3deff01028b6c4fbb", "generation_seconds_batch": 17.589386163977906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6883432865142822, "target_z": 0.75, "ce": 5.780362129211426, "genuineness": 4.7081804275512695, "identity": 0.6504747867584229, "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/emo_Pain/5c723e8c570bd34e8b8246e103c8f7d38a72933d0f3728f61b89d5fc80cbba1a.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": "5c9325726723e59b8ccb9a7c30c350ba1abf9cab921ba533e245ef31c81797da", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "793e2907adba67211c5861401853929071e4588be60ed25fadcea68afaf12821", "generation_seconds_batch": 15.6092861170182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44119924306869507, "target_z": 0.3203124983807918, "ce": 5.998785495758057, "genuineness": 4.3504462242126465, "identity": 0.5579948425292969, "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/emo_Pain/5c9325726723e59b8ccb9a7c30c350ba1abf9cab921ba533e245ef31c81797da.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "5ca231900c634ca1d696bba5b4a47d27fcec51883cc73e48e423b4faf69f1c04", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9acab8c3ef17b382ae6dd2d8792aa82f03369947f8563d606dde1011e67a7132", "generation_seconds_batch": 23.056892651016824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6050230264663696, "target_z": 2.3437500323841647, "ce": 5.863759994506836, "genuineness": 3.7040140628814697, "identity": 0.3063802123069763, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude graam. We kunnen er samen in kijken.", "wer": 0.05555555555555555, "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/emo_Pain/5ca231900c634ca1d696bba5b4a47d27fcec51883cc73e48e423b4faf69f1c04.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "5cc82dc762af6a3a4f43e02da4f9a5a9b1ca2c5db1008c12395163a19c6c2eb1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e759ba97eeafcf373c2025a5cc84c16e0ea3cb4100429c4854ceca303db1c9f5", "generation_seconds_batch": 23.084421940962784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11204831302165985, "target_z": -0.2519531278336144, "ce": 6.173524379730225, "genuineness": 4.595211505889893, "identity": 0.4786863923072815, "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/emo_Pain/5cc82dc762af6a3a4f43e02da4f9a5a9b1ca2c5db1008c12395163a19c6c2eb1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "5d2721e6c68a1df167b89c45796db4498fbeac2c2faf533d61b4889811edf8cb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "1896ea78e6d379ed69d56cfdb20c26171f14d0144fca36dcb7e49a5386f6eb45", "generation_seconds_batch": 24.054555406910367, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.204446479678154, "target_z": -0.09130860316164786, "ce": 5.824047088623047, "genuineness": 4.632867813110352, "identity": 0.6683136224746704, "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/emo_Pain/5d2721e6c68a1df167b89c45796db4498fbeac2c2faf533d61b4889811edf8cb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "5dec1b5a20b47f8598623544a5e2b8bae8118e9c6d8e58a1d61f7e5d67071a18", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "932bd64d2a04ed5486df68dd69747bd876259d3bf0616179a158cf221990f69a", "generation_seconds_batch": 21.51887546305079, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17327263951301575, "target_z": -0.14550780339195368, "ce": 6.125868320465088, "genuineness": 4.3881025314331055, "identity": 0.7275546193122864, "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/emo_Pain/5dec1b5a20b47f8598623544a5e2b8bae8118e9c6d8e58a1d61f7e5d67071a18.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "d72b4e4399c2e9cd9e73d9f1ab03e0035f276d218b3588768fb7f7eb68e8a305", "generation_seconds_batch": 17.164145344984718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04127524793148041, "target_z": -0.375, "ce": 6.109982967376709, "genuineness": 4.331617832183838, "identity": 0.8294896483421326, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/5e0af124ad7421917e897f2876d49b25e213a1bc87e01fe43f09b9c255c2a32f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "5ea1c92abf2fb5875728d9dae48833572ebd75d708d713c4b0a5c2c2976f6628", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "56bef47a9d18b9a35e55382879e13371634c2486458c3356d7c060d4d320a6e4", "generation_seconds_batch": 18.003477772930637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16147711873054504, "target_z": -0.16601563269123912, "ce": 6.419747352600098, "genuineness": 4.7834930419921875, "identity": 0.4730963706970215, "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/emo_Pain/5ea1c92abf2fb5875728d9dae48833572ebd75d708d713c4b0a5c2c2976f6628.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "5ee58bd770cd6661e59378c6be1fd4ae5eca8a37fbebe5aba1f3a306fca58ac6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2e5c4b8392f61a15f81093d284bf8dce0c10d842435b979a1c23bfc97b488521", "generation_seconds_batch": 17.57663396105636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5872389078140259, "target_z": 0.5742187540480206, "ce": 5.736677646636963, "genuineness": 4.68307638168335, "identity": 0.5096712112426758, "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/emo_Pain/5ee58bd770cd6661e59378c6be1fd4ae5eca8a37fbebe5aba1f3a306fca58ac6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "5f0efd9c3d7a29c224523c1843da98f9accdf467ea5ce1a099e51f1bf4c511be", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d1730f2e18f5aea0af18c3bdf7d323e85b61e6ab961207d35fd88724e78248a9", "generation_seconds_batch": 23.2788880179869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.062326908111572, "genuineness": 4.532451152801514, "identity": 0.5681689381599426, "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/emo_Pain/5f0efd9c3d7a29c224523c1843da98f9accdf467ea5ce1a099e51f1bf4c511be.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": "5f1665a3fa7024c5722323b3c9f194bd681f047f23398aaee8f1f594ee47a434", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9c3be84be6164ef9c981722cf3a9ea71500060491befa8f3c84bbd3d650b5df3", "generation_seconds_batch": 23.498963849968277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7287850379943848, "target_z": 0.8203124983807918, "ce": 5.200547218322754, "genuineness": 4.111956596374512, "identity": 0.509535551071167, "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/emo_Pain/5f1665a3fa7024c5722323b3c9f194bd681f047f23398aaee8f1f594ee47a434.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "5f5615a6d5320019e9c44f00a6f2eff7101d7e74e2b827950d62a32b55fd3a48", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "4f56f9b8f8a7b76bbe98a2a4f3a7ad43be392ec006e2c33626093c620f2313f7", "generation_seconds_batch": 28.277724315063097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.060372740030288696, "target_z": -0.3417968786432185, "ce": 6.149696350097656, "genuineness": 4.7332844734191895, "identity": 0.7546790242195129, "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/emo_Pain/5f5615a6d5320019e9c44f00a6f2eff7101d7e74e2b827950d62a32b55fd3a48.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "5facd83570d2eb09916d6e9c803c154f42d87e6edf77aeb6f76d7e7f955e4094", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "261a4c4132d03e4a0328056cf6a5a1868aa17113c71cebc991a2abff10dd41e0", "generation_seconds_batch": 21.79103117901832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11092492938041687, "target_z": -0.2539062556672288, "ce": 5.3792572021484375, "genuineness": 4.532451152801514, "identity": 0.4598411023616791, "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/emo_Pain/5facd83570d2eb09916d6e9c803c154f42d87e6edf77aeb6f76d7e7f955e4094.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "60d9856fc6984a6e9259f1a146b1493859fbec441f3b4aa48ded1510ee22f2ae", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b2adbb7eacee51c6edc27400100bad97f0dda98ebf41ee17af8756a8e9466abf", "generation_seconds_batch": 23.033237815019675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06935979425907135, "target_z": -0.326171881881635, "ce": 6.006728172302246, "genuineness": 4.632867813110352, "identity": 0.6595441102981567, "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/emo_Pain/60d9856fc6984a6e9259f1a146b1493859fbec441f3b4aa48ded1510ee22f2ae.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "6117e948bf91e5f6e2a85f0c4247d725049b3bf6078bb555e615d8651c884564", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ba70e341240ed0b5c0eadd6f5620b79c0f78a8d9695860a704c92cacf819e384", "generation_seconds_batch": 21.857697119936347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2849087119102478, "target_z": 0.04858396742391379, "ce": 5.879645347595215, "genuineness": 4.344170093536377, "identity": 0.5723400712013245, "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/emo_Pain/6117e948bf91e5f6e2a85f0c4247d725049b3bf6078bb555e615d8651c884564.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "6170e2b4e9d9379a032b40c6925065de217d2af452356c57c51f0f2f2ba11a83", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "23613fe4e93c04134a8e5402e9071e3b46df68b9a13ef39d28638dc3cf48c9ba", "generation_seconds_batch": 16.78800477704499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0907040536403656, "target_z": -0.2890625048576247, "ce": 6.117925643920898, "genuineness": 4.218649387359619, "identity": 0.3765583634376526, "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/emo_Pain/6170e2b4e9d9379a032b40c6925065de217d2af452356c57c51f0f2f2ba11a83.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "617d7d8db2f53148d64727355a5b57076a7797431daaccd779013131ba929907", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a06ab1912b192fb7b9909f37929f6e17615de02c8b71bcec6dd85420b80957c1", "generation_seconds_batch": 25.578460563905537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19911041855812073, "target_z": -0.10058594094081749, "ce": 5.982900142669678, "genuineness": 4.394378662109375, "identity": 0.5846424102783203, "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/emo_Pain/617d7d8db2f53148d64727355a5b57076a7797431daaccd779013131ba929907.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "61aea8f54eee59558dd74f7338f0ffafedf954dc1612164545637cf8f12d3e04", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7a818f842b27bf3f640fa74c68cc06905a3538e6afc5c262428bd8d757b3042d", "generation_seconds_batch": 23.458835119963624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4702171087265015, "target_z": 2.1093748996090893, "ce": 5.895530700683594, "genuineness": 4.055472373962402, "identity": 0.6727199554443359, "hypothesis": "The door's still open. Hey, put your coat on the chair. Tell me when you are ready.", "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/emo_Pain/61aea8f54eee59558dd74f7338f0ffafedf954dc1612164545637cf8f12d3e04.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": "61ce60cfa050d8ad717841013755d131f83a4bd1bc49b6ad0271ca679858f11f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5e778b9cabcf05880a3061262d813730e11548fa68d7de2548f3565a28567d77", "generation_seconds_batch": 21.618244612007402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6319841146469116, "target_z": 2.3906248931322565, "ce": 6.046441555023193, "genuineness": 3.9801597595214844, "identity": 0.3428972363471985, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verblaat. Het stond hier naast de hoge boekenkast. We zetten het na de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/61ce60cfa050d8ad717841013755d131f83a4bd1bc49b6ad0271ca679858f11f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "621f927792253a6060fca9de394649649977753c3355a666ce3f6475f121038c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "956356f04395e27146624bdd481aa75866ad51622aa78550644c699a2b848f20", "generation_seconds_batch": 21.492151062004268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04689215123653412, "target_z": -0.3652343867392597, "ce": 6.300607204437256, "genuineness": 4.68307638168335, "identity": 0.7298843860626221, "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/emo_Pain/621f927792253a6060fca9de394649649977753c3355a666ce3f6475f121038c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "622cf046ef79fa250f81936fb4815f769952c936a9a779c9d6a918f1f5931b25", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4c05c27aea0924869ee5a5d788a618b368c44ef4690fa84acc6476772b3cbb28", "generation_seconds_batch": 23.836500382982194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.058125972747802734, "target_z": -0.34570313431044736, "ce": 6.2767791748046875, "genuineness": 4.532451152801514, "identity": 0.7199912667274475, "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/emo_Pain/622cf046ef79fa250f81936fb4815f769952c936a9a779c9d6a918f1f5931b25.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": "6251953dd8c6765939fd2b8a21579ab4158689e697bb9a1a33d08dfda2811733", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "468e4d04abd086d7e3406bd3141fc25b6d957f7808aa9777bb08428c07b125c8", "generation_seconds_batch": 22.944967439048924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0972543954849243, "target_z": 1.4609375987717024, "ce": 5.601652145385742, "genuineness": 4.074300289154053, "identity": 0.04355160519480705, "hypothesis": "De sleutels liggen op de keukentafel. Ik klik 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/emo_Pain/6251953dd8c6765939fd2b8a21579ab4158689e697bb9a1a33d08dfda2811733.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": "6263770b4cd17a77823cfa5e38e97f4fcdbe421eb058ace046a76e10a6b0d6fd", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e6139499f9c780fbf355b7c3ca4692f176aa4fc4aad056b618ad54d61ff10dca", "generation_seconds_batch": 22.70014461502433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3097635507583618, "target_z": 0.09179687864321853, "ce": 6.141753673553467, "genuineness": 4.300237655639648, "identity": 0.5884571075439453, "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/emo_Pain/6263770b4cd17a77823cfa5e38e97f4fcdbe421eb058ace046a76e10a6b0d6fd.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": "62886603e3ae5c1e394f1f888c29863c137b70c01fc4eda3da085b3bf377b345", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b9718e46a0391e153f9bdaddf6ac8b528916563789241dd66328c44ef98cdd76", "generation_seconds_batch": 17.459654210018925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09407420456409454, "target_z": -0.2832031213567815, "ce": 6.141753673553467, "genuineness": 3.9299514293670654, "identity": 0.5316653847694397, "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/emo_Pain/62886603e3ae5c1e394f1f888c29863c137b70c01fc4eda3da085b3bf377b345.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": "63566360750fa31a5f63d62e897dc0b068e1aaa9369b83f165d0e644628aeac7", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "811bb4ff2a4744f70aeb541a1ae74197478a00c5fa767fd728192f14c4536583", "generation_seconds_batch": 16.8346429169178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7737202644348145, "target_z": 0.8984374044667142, "ce": 5.816103935241699, "genuineness": 4.293961524963379, "identity": 0.6859742403030396, "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/emo_Pain/63566360750fa31a5f63d62e897dc0b068e1aaa9369b83f165d0e644628aeac7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "6398aa624077af1223d0c97405bc013e94b37011d42a076d3b386b30c9661753", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "99b6868156fd82cbc9816fc452e33338f2fcbc8331ed07cbfeb3ceb36464dda1", "generation_seconds_batch": 23.033237815019675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1513666808605194, "target_z": -0.18359375728643706, "ce": 5.895530700683594, "genuineness": 4.432034492492676, "identity": 0.3532260060310364, "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/emo_Pain/6398aa624077af1223d0c97405bc013e94b37011d42a076d3b386b30c9661753.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": "63b3c3ad99078728fe58f07cc33270d406f34da7d4b5e4c12ae49a0069d27e03", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "dd5d14badffe1f936b3214a0941de65d79a7b22d4eeb25d1cbb93eadcc035885", "generation_seconds_batch": 15.234316315036267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5241395235061646, "target_z": 2.2031250356225813, "ce": 5.903473377227783, "genuineness": 4.419482707977295, "identity": 0.4628611207008362, "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": 0.03999999999999915, "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/emo_Pain/63b3c3ad99078728fe58f07cc33270d406f34da7d4b5e4c12ae49a0069d27e03.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "657c5c9053aee5aefea0387438691a0d78cc50f84bc9790ca9e53fa4cdee7653", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "654635767c199c032628e79eff64510b1433db1f1eec99836a61b7510be12b08", "generation_seconds_batch": 22.28077194094658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4277186393737793, "target_z": 0.29687496437741884, "ce": 6.1894097328186035, "genuineness": 4.8588056564331055, "identity": 0.8037810921669006, "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/emo_Pain/657c5c9053aee5aefea0387438691a0d78cc50f84bc9790ca9e53fa4cdee7653.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "65861000465719cb1bdaef738b7e358b16e933ed478eec0b05e99506a60be4f2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2e0f7f47ef5908cbe867d79b51f6e4a08cc2de1d54c265e1731a00059deef80a", "generation_seconds_batch": 22.3828182220459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 5.95112943649292, "genuineness": 4.557555198669434, "identity": 0.5262320637702942, "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/emo_Pain/65861000465719cb1bdaef738b7e358b16e933ed478eec0b05e99506a60be4f2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "659306032a7834b7791a2d32f21c61501dc83ad6130f63fd78e4b2d7aec18681", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "44b5638c842295a12f260436cf28cbd9c3475c792f0d426d391f8a2195e5fe4e", "generation_seconds_batch": 16.81523526192177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15361344814300537, "target_z": -0.17968750161920824, "ce": 5.895530700683594, "genuineness": 4.482243061065674, "identity": 0.4449613392353058, "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/emo_Pain/659306032a7834b7791a2d32f21c61501dc83ad6130f63fd78e4b2d7aec18681.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "65c71ce9b28567770199d2ad96537c39403f2c8b3c3cd851cadb80f4470d5a09", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ffa03032e5ff46c44bdb8e01a3b6cc8e1876322529395c3da16f33b561fb1273", "generation_seconds_batch": 21.739319278975017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 6.022613525390625, "genuineness": 4.607763767242432, "identity": 0.5609281063079834, "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/emo_Pain/65c71ce9b28567770199d2ad96537c39403f2c8b3c3cd851cadb80f4470d5a09.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "65d8cd742e61f6175e4177a81180c6ba24806ee96a54b77ceae194c2cb6dff3d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8024ceb9b7d6205efe2d8a81c8ee992c67e41fa9837c3242f222f28eaeb3eeba", "generation_seconds_batch": 17.75086737505626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43670570850372314, "target_z": 0.31249998704633414, "ce": 5.927301406860352, "genuineness": 4.808597087860107, "identity": 0.7638697624206543, "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/emo_Pain/65d8cd742e61f6175e4177a81180c6ba24806ee96a54b77ceae194c2cb6dff3d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "65f2b883a7a7ab824710f34eaa4e433a29e92e58f2ea1b13a56574172341e5e9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5873a9148f8ebe998eb97f1dadfa4223a0cbd760a863a4d2e990295280b2532e", "generation_seconds_batch": 22.604677945026197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15305176377296448, "target_z": -0.18066405258234955, "ce": 6.109982967376709, "genuineness": 4.720732688903809, "identity": 0.7649359703063965, "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/emo_Pain/65f2b883a7a7ab824710f34eaa4e433a29e92e58f2ea1b13a56574172341e5e9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "65fb0249a216c86a6d15f9d60301d880dcba8ce83e3468df9a9cfd9815f168eb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5cf897e90bc8c8c2230ddbbaec78cc459334bcd276ab42640e23b63f6f7c48fb", "generation_seconds_batch": 17.430287464987487, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.09409761428833, "genuineness": 4.796044826507568, "identity": 0.6973830461502075, "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/emo_Pain/65fb0249a216c86a6d15f9d60301d880dcba8ce83e3468df9a9cfd9815f168eb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "660382c8c43227874e0756ad8c6bc70335148c711559929e068b6ff3801e0a4b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6025247cb1aece6902793af5f96b0d45fd412a34bbc578a72d7acc1a280e5131", "generation_seconds_batch": 15.275349429924972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1691508293151855, "target_z": 1.5859375728643705, "ce": 5.959072113037109, "genuineness": 4.281409740447998, "identity": 0.11396391689777374, "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/emo_Pain/660382c8c43227874e0756ad8c6bc70335148c711559929e068b6ff3801e0a4b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "661d1559bb261f704398244683d73da30e8552b3b1948999fd529435fa86eff6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "1ec3155a0e1c4b89c697db8e15a8b3425cc066c5be551e4698813f277ceb5e06", "generation_seconds_batch": 22.15373884292785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19630196690559387, "target_z": -0.10546874757118765, "ce": 6.117925643920898, "genuineness": 4.582659721374512, "identity": 0.42370766401290894, "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/emo_Pain/661d1559bb261f704398244683d73da30e8552b3b1948999fd529435fa86eff6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "66919afa58574eb2051c0928844f81e5b4e0bbf7abac972aa93650ebedba34d5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "da82fef187e48daacf6f5620adf4d6b463ae53e371247a19b4d75b5211ef8b32", "generation_seconds_batch": 16.45040664600674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08621053397655487, "target_z": -0.2968749902847506, "ce": 6.173524379730225, "genuineness": 4.770940780639648, "identity": 0.7930195927619934, "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/emo_Pain/66919afa58574eb2051c0928844f81e5b4e0bbf7abac972aa93650ebedba34d5.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": "66f0cca6a4b1ddef970aa774ca38daf4e24ebf01af6e867facee46cd6931d020", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Natural clear speech.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Natural clear speech.) [3.2 seconds duration] Ruf an, wenn du Zeit hast.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "66e765a7dc14a3cddbd3a65137e1459dd0766990d56803d166faa93148277a74", "generation_seconds_batch": 23.868913928978145, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24320316314697266, "target_z": -0.023925783577611836, "ce": 6.1894097328186035, "genuineness": 4.808597087860107, "identity": 0.6744820475578308, "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/emo_Pain/66f0cca6a4b1ddef970aa774ca38daf4e24ebf01af6e867facee46cd6931d020.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "677146e8877bf1ce9f8b97201b4d18e2ceacf9c76629aeb46f69087139a7a7c2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a4cb26e09e4d7c736059a8c677f1eac3781aa7377af0934840e41de374bb353b", "generation_seconds_batch": 22.949783375952393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1148567646741867, "target_z": -0.24707032120324426, "ce": 6.062326908111572, "genuineness": 4.406930446624756, "identity": 0.6536257863044739, "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/emo_Pain/677146e8877bf1ce9f8b97201b4d18e2ceacf9c76629aeb46f69087139a7a7c2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "677d5b0f422c2835b5232acaf0f4245209059c93bc26d74b4e93bb9977a8c633", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6cdcf77898c98e71eb66997b838bf4e4d564530d1c5727362e043feedd64e947", "generation_seconds_batch": 22.915941952029243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05138568580150604, "target_z": -0.3574218754048021, "ce": 6.316492557525635, "genuineness": 4.3881025314331055, "identity": 0.5576229095458984, "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/emo_Pain/677d5b0f422c2835b5232acaf0f4245209059c93bc26d74b4e93bb9977a8c633.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "683bc5029c09e81a9d7775b5c861457052b8eef5c45a0dacfd8fc4f2ed16ff46", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dd88f35d952c8d4fba038c4aae8582630f40e6d0fdcfa5947f65f1a875ef0ca6", "generation_seconds_batch": 17.146380318910815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.173524379730225, "genuineness": 4.758388519287109, "identity": 0.672476589679718, "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/emo_Pain/683bc5029c09e81a9d7775b5c861457052b8eef5c45a0dacfd8fc4f2ed16ff46.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "68d50551a5d05881a5634dd4e0d525e95e6bc09a9b663af9ae233f111f5ad9ce", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b83c9272ae9341ad5e7dec151904eb986bbbc86fc9fc86c4b63e99030f6a0278", "generation_seconds_batch": 16.6087052320363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14125624299049377, "target_z": -0.20117188188163498, "ce": 6.23706579208374, "genuineness": 4.494794845581055, "identity": 0.39360150694847107, "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/emo_Pain/68d50551a5d05881a5634dd4e0d525e95e6bc09a9b663af9ae233f111f5ad9ce.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "69dd552d3dbad99272b1522165b2fd7510919242df8c8b039d4e105e3667ccdf", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "ad11820cc2ec58ded5a92d5a33251b08ab6a5baff244ff1fcf97081f898230c7", "generation_seconds_batch": 24.454157738015056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.757802963256836, "target_z": 2.6093750032384166, "ce": 5.7406487464904785, "genuineness": 4.432034492492676, "identity": 0.6383096575737, "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/emo_Pain/69dd552d3dbad99272b1522165b2fd7510919242df8c8b039d4e105e3667ccdf.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "69e4f72a5dd48d5e2b8250905b3f51b96aaf2c99e1a16387b577d4c25aa225d7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1702, "scale": 0.0, "cfg": 1.5, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "80f36cb81e096bc2b2bb9762a749718bc50799389f31f17e3c11356b8aabc1df", "generation_seconds_batch": 17.062629845924675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09744435548782349, "target_z": -0.27734373785593824, "ce": 6.157639026641846, "genuineness": 4.444586753845215, "identity": 0.4388582408428192, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/69e4f72a5dd48d5e2b8250905b3f51b96aaf2c99e1a16387b577d4c25aa225d7.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "b283bcf9f7abf745820dce2c1725f465db09f736e1cbaa7cf362856056bcec23", "generation_seconds_batch": 17.480708557064645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04464539885520935, "target_z": -0.36914061649915675, "ce": 6.459460258483887, "genuineness": 4.143336772918701, "identity": 0.214301198720932, "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/emo_Pain/6a61bc20d43b3e97640d8dd45682f7055c5dbaa5aa987cdb55f8a7d7a432a2c2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "6a8180144ac280d0adf3db5fe6e2e2820d470dbfb90c0bf234cd0ccc7b7554a4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "39205406e7a71e712af5af960fcdc2ee97f60f505d43a093cf3254efccaff166", "generation_seconds_batch": 23.037476580007933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6411612033843994, "target_z": 0.6679686828028583, "ce": 5.696964263916016, "genuineness": 4.720732688903809, "identity": 0.6981178522109985, "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/emo_Pain/6a8180144ac280d0adf3db5fe6e2e2820d470dbfb90c0bf234cd0ccc7b7554a4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "6a97197f3a53b9aa4b14c9516b30f7ba01ac9257e0dd59ab3cdd7b3f4d408d71", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fcaf6df81b783124d940cfb295c4077ae6e0084bdcecff81b91dea9b9a9d1d4e", "generation_seconds_batch": 21.106439461000264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24573077261447906, "target_z": -0.019531252428812353, "ce": 5.927301406860352, "genuineness": 4.3127899169921875, "identity": 0.6309065222740173, "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/emo_Pain/6a97197f3a53b9aa4b14c9516b30f7ba01ac9257e0dd59ab3cdd7b3f4d408d71.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "6abfc361af82b8c56e64e607a421babbc9e5c2188b0f70d6d07fbf300b87582f", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ffc93e030dbcf1a460c3679d7f3db27cd713353591bc04288daf04f718832186", "generation_seconds_batch": 16.934556304011494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.037905097007751465, "target_z": -0.38085938350084325, "ce": 6.332377910614014, "genuineness": 4.394378662109375, "identity": 0.5616929531097412, "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/emo_Pain/6abfc361af82b8c56e64e607a421babbc9e5c2188b0f70d6d07fbf300b87582f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "6b0608c46b6033324621e9f4d19d66a5d2ca7c41ee0287d7a13d7d35d223ad65", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5d90156105dd88effc874d3176ac062245a3eabf82fc851c83e9a6ede274705c", "generation_seconds_batch": 22.930587901035324, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8296993970870972, "target_z": 2.734374977331085, "ce": 6.133810997009277, "genuineness": 4.381826400756836, "identity": 0.36654090881347656, "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/emo_Pain/6b0608c46b6033324621e9f4d19d66a5d2ca7c41ee0287d7a13d7d35d223ad65.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "6b0a41970951a28b68d0fdfbfc69e2c4e1e77708781ebbeb6af1119ebfdc0b3c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "af0688b1e02e4db492035657751f4a9c898fd5aad1c02ecb4d7ddada7ce48259", "generation_seconds_batch": 22.44372973893769, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7175512313842773, "target_z": 0.8007812718593111, "ce": 5.677107334136963, "genuineness": 4.482243061065674, "identity": 0.35965102910995483, "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/emo_Pain/6b0a41970951a28b68d0fdfbfc69e2c4e1e77708781ebbeb6af1119ebfdc0b3c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "6b0b3eb9081f6968ed474f9cfef4348c8dc33fe4be69c9e707c45183295c2678", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b19cbc0b7022994a31d85654ff16383aa823d50c1a863ccaca373739f9aa1e7b", "generation_seconds_batch": 22.627415099996142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09744435548782349, "target_z": -0.27734373785593824, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.2631717622280121, "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/emo_Pain/6b0b3eb9081f6968ed474f9cfef4348c8dc33fe4be69c9e707c45183295c2678.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": "6b96842191691661b936a092fa5bec3aa816bcdf49af6140dc78a78dddb35e0b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] La colocaremos después de comer.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Natural clear speech.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Natural clear speech.) [3.2 seconds duration] La colocaremos después de comer.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fef614817f2a405b8ca205cb467f8650ecc728119348e0772fbe4effc98f132", "generation_seconds_batch": 17.007002598023973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4895046651363373, "target_z": 0.40429686568955264, "ce": 5.95112943649292, "genuineness": 4.607763767242432, "identity": 0.6394326686859131, "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/emo_Pain/6b96842191691661b936a092fa5bec3aa816bcdf49af6140dc78a78dddb35e0b.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "98b7bc0e16632f7939e452d606168fcd309bcac9b338873deb3b905fcd8b5612", "generation_seconds_batch": 17.45467336208094, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.060372740030288696, "target_z": -0.3417968786432185, "ce": 6.006728172302246, "genuineness": 4.796044826507568, "identity": 0.6533839702606201, "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/emo_Pain/6b973c750e3616c70f28663c29496a0c74bea4a87dff4bad65cadc5394b99131.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "6c55942053cbbac90576a7b1f04ef0dad7f69b4d40bd60300cdc158f29903e43", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "58f08530ab79cbe3638677511db6362efe55b896e1d4228c89c47b839aebdfaf", "generation_seconds_batch": 21.70856671698857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.9645051956176758, "target_z": 2.968749902847506, "ce": 5.363371849060059, "genuineness": 4.174716949462891, "identity": 0.5422770380973816, "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/emo_Pain/6c55942053cbbac90576a7b1f04ef0dad7f69b4d40bd60300cdc158f29903e43.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "6c6790961f2ac6e91586414b3f4a1848a5e085028d949e0b84066417ea8fe222", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "08ea27b8cd96b59835e9aa1343f00877770281585cbdaec01ea90e57c8c1e031", "generation_seconds_batch": 23.933782907901332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11373338103294373, "target_z": -0.24902344903685866, "ce": 6.316492557525635, "genuineness": 4.645420074462891, "identity": 0.5546020269393921, "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/emo_Pain/6c6790961f2ac6e91586414b3f4a1848a5e085028d949e0b84066417ea8fe222.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "6cfee05c8bc24d9391699b52c892bd6c5ae8b71698155b8423616b675c4d2864", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "af6b6e69eb9c10ce71c9d452d63b525e123603fe2133187c2ae9a727b217cf9e", "generation_seconds_batch": 24.89237240003422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3546988368034363, "target_z": 0.16992188835846794, "ce": 5.839932441711426, "genuineness": 4.7458367347717285, "identity": 0.7413454055786133, "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/emo_Pain/6cfee05c8bc24d9391699b52c892bd6c5ae8b71698155b8423616b675c4d2864.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "6d02d0192503e5233cd47a0913f1c1cdcce1fac3fe50695522b01afc1f51c002", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9a1167923c7eea37824e6751237defa55d598bca4723fcbc4e5ee8f32d37e372", "generation_seconds_batch": 23.27977334696334, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23393526673316956, "target_z": -0.040039055820766035, "ce": 5.764476776123047, "genuineness": 4.344170093536377, "identity": 0.3460835814476013, "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/emo_Pain/6d02d0192503e5233cd47a0913f1c1cdcce1fac3fe50695522b01afc1f51c002.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "6f7ee988e9a0f3470ee24af185856e6c5d97a8f67cd9f07a68eb68bcc68be12e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "972a3edec1761c96d06e48647d0dac5e8611a144b3bac9023863712018a1dc02", "generation_seconds_batch": 23.261758264969103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3906470537185669, "target_z": 0.23242187540480205, "ce": 5.974957466125488, "genuineness": 4.720732688903809, "identity": 0.594465434551239, "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/emo_Pain/6f7ee988e9a0f3470ee24af185856e6c5d97a8f67cd9f07a68eb68bcc68be12e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "6f99ab63c41be7f13f69eb5ef69aa671e0ccdf20b30f80b7491920bf32e678ba", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d68fb43adda81fb02a5269246dacc1e93cda39f29968bb4de417d61e73249051", "generation_seconds_batch": 23.309764986042865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.49062806367874146, "target_z": 0.4062500194304988, "ce": 5.80021858215332, "genuineness": 4.595211505889893, "identity": 0.7989786863327026, "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/emo_Pain/6f99ab63c41be7f13f69eb5ef69aa671e0ccdf20b30f80b7491920bf32e678ba.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "707a22882a6a7c6f203a95ce854ad5e7a1ebcbb09a033dc6115a7ad22e01d7d9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d993b917c1604be0bf2cedf5408ffbe6f77a0bc2892824db795f3332926aff7c", "generation_seconds_batch": 20.67807626002468, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05138568580150604, "target_z": -0.3574218754048021, "ce": 6.221180438995361, "genuineness": 4.545003414154053, "identity": 0.5092625617980957, "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/emo_Pain/707a22882a6a7c6f203a95ce854ad5e7a1ebcbb09a033dc6115a7ad22e01d7d9.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": "7127ab0196dccb48e37150c91fc6d756adc1023ea9b87bd1a77416cd92105df4", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f6492efe65c731674ec46048067b6941cfb6ff8487ba7374ace790185d61891e", "generation_seconds_batch": 21.43925783596933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3072359263896942, "target_z": 0.08740232158708729, "ce": 5.661221981048584, "genuineness": 4.369274139404297, "identity": 0.732054591178894, "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/emo_Pain/7127ab0196dccb48e37150c91fc6d756adc1023ea9b87bd1a77416cd92105df4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "712ecaf0afada2731f0b73507be84e4932e2e255123c60eda61b854aee5b13df", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2beed24ec1d362acba6904af41702ce53bce47bec0af74d98a8c940eb5d82cd3", "generation_seconds_batch": 22.002731401007622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.469283789396286, "target_z": 0.36914061649915675, "ce": 6.086154937744141, "genuineness": 4.268857479095459, "identity": 0.6530855894088745, "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/emo_Pain/712ecaf0afada2731f0b73507be84e4932e2e255123c60eda61b854aee5b13df.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "714d71ceed027bc65eb7ff806fa25babdb12a5d7839da61cc047d1ada1f6be31", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4924082dfe9f99d27fa099aa69bcad4699571293cdc69efd1c21c5ae05480363", "generation_seconds_batch": 16.333416844019666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4445694088935852, "target_z": 0.32617190778896676, "ce": 5.784333229064941, "genuineness": 4.595211505889893, "identity": 0.45923927426338196, "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/emo_Pain/714d71ceed027bc65eb7ff806fa25babdb12a5d7839da61cc047d1ada1f6be31.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": "715545f91fd5ee5e2076f794f6823b7dee0d0b8c781966b29791654422e4d4cd", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e2eef99c49bcf1a9f3b5becc3fb98b273f194220e62fa22e52d2f9b613ae8e6e", "generation_seconds_batch": 25.250903810025193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4142380952835083, "target_z": 0.27343753400337295, "ce": 5.657250881195068, "genuineness": 4.15588903427124, "identity": 0.7319059371948242, "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/emo_Pain/715545f91fd5ee5e2076f794f6823b7dee0d0b8c781966b29791654422e4d4cd.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "723a60863a569228fda76be2dcb5a0b2c096531b5e7e3404f0049da6a69c9aff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "435bd6e301c8c57aca53235d2f64091e493558764491330d9be917a4d9ac1997", "generation_seconds_batch": 25.245366242947057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.757802963256836, "target_z": 2.6093750032384166, "ce": 5.510311126708984, "genuineness": 4.143336772918701, "identity": 0.24231550097465515, "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/emo_Pain/723a60863a569228fda76be2dcb5a0b2c096531b5e7e3404f0049da6a69c9aff.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "72aa38f71940f27f985f6be26ac4b3d4b8b895aab13807c37841dc9c24665d30", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "004d22fbda480da32c998ca1b70be068059f53a928114ea8b8cfdc666d87cbe9", "generation_seconds_batch": 23.28106885298621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.23706579208374, "genuineness": 4.7458367347717285, "identity": 0.49073660373687744, "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/emo_Pain/72aa38f71940f27f985f6be26ac4b3d4b8b895aab13807c37841dc9c24665d30.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "72e9f62080e7f7ab1910a5ceb5624875f26e997548b44f032621ffcf3f0feed2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7e58d2adaf0754f376b47fcb9fac5d4ac76744c82f19ae0a73da86e6f76524a4", "generation_seconds_batch": 23.32275474607013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.443574905395508, "genuineness": 4.7834930419921875, "identity": 0.6854647397994995, "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/emo_Pain/72e9f62080e7f7ab1910a5ceb5624875f26e997548b44f032621ffcf3f0feed2.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": "730a4b520512bc13fe8a6c94f248b0c2a18ead0614021b8318bd97d22d257491", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a3bb19db5ac87d50fac26be40e91988523af3587681f570a865944ee04b99a58", "generation_seconds_batch": 18.38001204095781, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1390094757080078, "target_z": -0.2050781375488638, "ce": 6.181467056274414, "genuineness": 4.557555198669434, "identity": 0.4690742790699005, "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/emo_Pain/730a4b520512bc13fe8a6c94f248b0c2a18ead0614021b8318bd97d22d257491.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": "735d744ca6eea75adcabd2babee685fa071baef1f7f9bf9ae858f4dc3f40a280", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "22965b8ca2d72a1395ce4ed60796004651e02267d6480a9023ca57bd6bcbc0a0", "generation_seconds_batch": 16.215985237038694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6523950695991516, "target_z": 0.6875000129536659, "ce": 5.95112943649292, "genuineness": 4.582659721374512, "identity": 0.8188709020614624, "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/emo_Pain/735d744ca6eea75adcabd2babee685fa071baef1f7f9bf9ae858f4dc3f40a280.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "7383337f72a9552e9560c8af6daa42fd74ebea77b85c844317d083915e8327da", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "90863dce55d0839fd54e91a12d99215c13244ccccf532412676427d4796418cf", "generation_seconds_batch": 22.52575620589778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08621053397655487, "target_z": -0.2968749902847506, "ce": 5.855817794799805, "genuineness": 4.381826400756836, "identity": 0.38350868225097656, "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/emo_Pain/7383337f72a9552e9560c8af6daa42fd74ebea77b85c844317d083915e8327da.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "738b6ef4ce2524efe7b8a0caefe3b6d90949924038640486e26925b9e84ebf51", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8fa189b1f4e7cdefc23fe6097dbaf41fea8e8e18353f2dc4374809d3b86165a8", "generation_seconds_batch": 17.589386163977906, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0433319807052612, "target_z": 1.3671874627582106, "ce": 5.8875885009765625, "genuineness": 4.7332844734191895, "identity": 0.7100249528884888, "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/emo_Pain/738b6ef4ce2524efe7b8a0caefe3b6d90949924038640486e26925b9e84ebf51.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": "73addef5853206169204c80364d40cfc8c4b354eaf4d81b47d16274290f7d122", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e8e6c6d5f99b4dec44d2f3cd6cbdb86acd672ee255d7eef2152b324328bbd153", "generation_seconds_batch": 22.583702816977166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7647333145141602, "target_z": 0.8828125890564529, "ce": 5.863759994506836, "genuineness": 3.9425034523010254, "identity": 0.5572632551193237, "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/emo_Pain/73addef5853206169204c80364d40cfc8c4b354eaf4d81b47d16274290f7d122.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "73be2e8654bb4efb15868ce768773a873c0c98752015051ae09ef29871a85f4a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "53cdae46311f1686d3fd487c5f333f6d4658b749b128797242e0fabc02ac3fb8", "generation_seconds_batch": 23.359805307001807, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33335456252098083, "target_z": 0.1328124854271259, "ce": 5.363371849060059, "genuineness": 4.519899368286133, "identity": 0.6447951793670654, "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/emo_Pain/73be2e8654bb4efb15868ce768773a873c0c98752015051ae09ef29871a85f4a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "74918c0cf7cf0e189fe2027de1b8424ceedd19ab5147eb54b26c3b1e34101106", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f15e6e5a53c3ffebf470d5c2bae248b962b396b8e70b8c531b5e09a35c66a0f7", "generation_seconds_batch": 22.74764752702322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10755479335784912, "target_z": -0.2597656132607403, "ce": 6.260893821716309, "genuineness": 4.557555198669434, "identity": 0.45670202374458313, "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/emo_Pain/74918c0cf7cf0e189fe2027de1b8424ceedd19ab5147eb54b26c3b1e34101106.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": "74b296af1e0506f0a0a6a4a00dd0917681c6987918b74c76aa6655e8194e740c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "d2c26a76c2a52a8bf1ea65b53c4326b8a5c25bb5c966042bffacde9f0d9b124a", "generation_seconds_batch": 22.557077716104686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5535374879837036, "target_z": 0.5156250744835787, "ce": 6.062326908111572, "genuineness": 4.030367851257324, "identity": 0.7826600074768066, "hypothesis": "I wrote the number on the envelope. Read it carefully. Check the last two digits. Call me when you have a moment.", "wer": 0.043478260869565216, "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/emo_Pain/74b296af1e0506f0a0a6a4a00dd0917681c6987918b74c76aa6655e8194e740c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "752914caf9dfb28502dddecce423063193a6e88d2fef41ced1429b4d28096a84", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7bc5fc044c6d55a07b78d613ba79fb4ac123a3b5b5035f0ffc50c86863cae64c", "generation_seconds_batch": 23.01767224597279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07160656154155731, "target_z": -0.3222656262144062, "ce": 6.149696350097656, "genuineness": 4.632867813110352, "identity": 0.7777226567268372, "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/emo_Pain/752914caf9dfb28502dddecce423063193a6e88d2fef41ced1429b4d28096a84.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "752cd5d9369f9f0b6fa0aa1573a5ee76d9eeffcfc01b17eccd7daabb34c180fc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "d25d0f8c61e7379dc28a5ed38049328294723366e4a42f33890dc20abaad6464", "generation_seconds_batch": 18.554241056088358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7108108997344971, "target_z": 0.7890624530429612, "ce": 6.0305562019348145, "genuineness": 3.8044304847717285, "identity": 0.4008956551551819, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/752cd5d9369f9f0b6fa0aa1573a5ee76d9eeffcfc01b17eccd7daabb34c180fc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "75764bfa881068ef7397590c977c28f4d489bd9155107d93dcf1c13aeb7b6c1d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "77deda65d76cfe2490368443d9f0a8658851eaf1e78c8f26ed921d9f9d4afac3", "generation_seconds_batch": 22.87817033799365, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9040325880050659, "target_z": 1.1249999222780047, "ce": 5.871703147888184, "genuineness": 4.469690799713135, "identity": 0.3377256691455841, "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/emo_Pain/75764bfa881068ef7397590c977c28f4d489bd9155107d93dcf1c13aeb7b6c1d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "75b2b3b1a4aaeaf5f3801642dc3c7f75111f08606ab265f209cda04d9b07e757", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "add4ce40e477a3269077ba0d526cfeef0fd5f180ac3ea405d0e5bbd7b6be2390", "generation_seconds_batch": 24.174109110957943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3758530616760254, "target_z": 1.94531247247346, "ce": 5.895530700683594, "genuineness": 4.369274139404297, "identity": 0.4234577417373657, "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/emo_Pain/75b2b3b1a4aaeaf5f3801642dc3c7f75111f08606ab265f209cda04d9b07e757.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "75d536137695db81a276f24f46bba842bff96f252681faca20131f90c057395d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "33702931d10f1d35b88a22be69d545b04259d6590a8753aaef430a477c2bdef1", "generation_seconds_batch": 18.31059727095999, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.189561665058136, "target_z": -0.11718751457287412, "ce": 6.09409761428833, "genuineness": 4.796044826507568, "identity": 0.7163760662078857, "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/emo_Pain/75d536137695db81a276f24f46bba842bff96f252681faca20131f90c057395d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "765af0240708ae6a4014081c798caae12b01916e8ee3a050f496168a05ba5999", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "04c38170335b71c1f65f1b618baa9bbfe2e6a9c2f9ba52e4d35f274e9183e5bf", "generation_seconds_batch": 22.669879650929943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19237013161182404, "target_z": -0.1123046820351722, "ce": 5.903473377227783, "genuineness": 4.65797233581543, "identity": 0.6185837388038635, "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/emo_Pain/765af0240708ae6a4014081c798caae12b01916e8ee3a050f496168a05ba5999.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "76741f0f5c48b7276facda8fad9ac90125de48276190c1d1ea85de9ddf1a68f5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6a9049badacaa60b0623221c20f2ae0ff8735c75a020931f07f1afd2f8b166b4", "generation_seconds_batch": 17.632916691945866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5187126398086548, "target_z": 0.45507818936352734, "ce": 6.1020402908325195, "genuineness": 4.7332844734191895, "identity": 0.6340039968490601, "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/emo_Pain/76741f0f5c48b7276facda8fad9ac90125de48276190c1d1ea85de9ddf1a68f5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "76dcfee38286a518f711d26aa5a04412c735b69b4fae64d110a92456b7727a6d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f52756c819f507aa9edc553ba8784dce5e34af89f2bfd5a5e4e6fb195f2f5b4c", "generation_seconds_batch": 24.323153220932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.25359445810317993, "target_z": -0.005859357593511478, "ce": 5.375286102294922, "genuineness": 4.607763767242432, "identity": 0.5630922317504883, "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/emo_Pain/76dcfee38286a518f711d26aa5a04412c735b69b4fae64d110a92456b7727a6d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "76f0656f02e83e64997287387d001ad80e26d1684af300ad5b14dfbdc24af840", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c6ce85c7ad5ce68c2c9e66a75dc549cf54781ffb6ca8dab738ce8b2410f3ae8f", "generation_seconds_batch": 21.688529993058182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19265097379684448, "target_z": -0.11181640655360155, "ce": 6.221180438995361, "genuineness": 4.319066047668457, "identity": 0.6846444606781006, "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/emo_Pain/76f0656f02e83e64997287387d001ad80e26d1684af300ad5b14dfbdc24af840.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "770bc6c0eac75724c905cbe3c993e847d9a180cd9803314eb100a82690b8a511", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d971e4f43cd4a2c6dbae4c21d67e46cf699d6fe094cc77e43a50b280594c0ce8", "generation_seconds_batch": 23.52213394199498, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12328213453292847, "target_z": -0.23242187540480205, "ce": 5.982900142669678, "genuineness": 4.7081804275512695, "identity": 0.6416534781455994, "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/emo_Pain/770bc6c0eac75724c905cbe3c993e847d9a180cd9803314eb100a82690b8a511.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": "776f375f53cae2bc35d0fecc48ca4d91d514d20a200a79ad25de9b463e7c4c55", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5fbfce20f7e631b7af0c9c995ec068bb9affa0071ea3952e949ada8ddb270ada", "generation_seconds_batch": 17.037321664043702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.629927396774292, "target_z": 0.6484374562813776, "ce": 5.895530700683594, "genuineness": 4.494794845581055, "identity": 0.8197475671768188, "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/emo_Pain/776f375f53cae2bc35d0fecc48ca4d91d514d20a200a79ad25de9b463e7c4c55.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "777d3ba2966d7ecb410bbf11542f30b7cbccb1b074fa69b8606f84f994987966", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3b936b21beaaaf4b59cc58b595e59c1e2c16f93336a6ea7b5ecb665e1ebc15df", "generation_seconds_batch": 23.7774008710403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8680844306945801, "target_z": 1.0625000388609975, "ce": 5.760505676269531, "genuineness": 4.557555198669434, "identity": 0.6111205816268921, "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/emo_Pain/777d3ba2966d7ecb410bbf11542f30b7cbccb1b074fa69b8606f84f994987966.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "779008987ed076ebf7c684552fd311d4338fd73b831b49ff10706950130593cd", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b64ae848cdc672463bdee07fc6a31a4f2f0119599b39265c7cdeab6fa83e178a", "generation_seconds_batch": 22.554827383020893, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3533854484558105, "target_z": 1.9062500194304988, "ce": 5.645336627960205, "genuineness": 4.400654315948486, "identity": 0.46553879976272583, "hypothesis": "La puerta sí abierta. He dejado tu abrigo sobre la silla. Dime cuándo estás listo.", "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/emo_Pain/779008987ed076ebf7c684552fd311d4338fd73b831b49ff10706950130593cd.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "2a1b1ed2378a7590023b87c76b496c861f842ae51d2c781bcd76bbf84a11b19d", "generation_seconds_batch": 16.218586139031686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.260893821716309, "genuineness": 4.6956281661987305, "identity": 0.764392077922821, "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/emo_Pain/77962898fb95ddc0e92a153772191fe01482e40f98d2245158cb92652e8a93c1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "77d026a08165ad5ee55c333cbc35ff68261d25ad9695067299ae9583e492d1bf", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "43205941173959f885f80039f986f07bfd00599e4ed7e885ab47c47b774184ee", "generation_seconds_batch": 24.103807280072942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7108108997344971, "target_z": 0.7890624530429612, "ce": 5.847874641418457, "genuineness": 4.758388519287109, "identity": 0.7109799385070801, "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/emo_Pain/77d026a08165ad5ee55c333cbc35ff68261d25ad9695067299ae9583e492d1bf.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "7819931139658961208c3100bf695cea9e95985e67334a58e48fc970ba1cb4fb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "922fb82b758d9388ffc784630f98fc62b120d84cd73cbe33f4e513e42c36405d", "generation_seconds_batch": 22.793283785926178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.27521955966949463, "target_z": 0.03173829491206948, "ce": 6.006728172302246, "genuineness": 4.607763767242432, "identity": 0.6730363965034485, "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/emo_Pain/7819931139658961208c3100bf695cea9e95985e67334a58e48fc970ba1cb4fb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "7849d02b2bf15fe5b7a1a07f6f6d4f73d76661e12674fe06f13f35d007560d17", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c06210e67c150915dbc35cbcac40b5bae25c000fd79034e8d9d920f62207bf45", "generation_seconds_batch": 16.399933042936027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09969110786914825, "target_z": -0.27343750809604117, "ce": 6.125868320465088, "genuineness": 4.482243061065674, "identity": 0.2858300805091858, "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/emo_Pain/7849d02b2bf15fe5b7a1a07f6f6d4f73d76661e12674fe06f13f35d007560d17.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "784bdf4713e2c0d3f82251f25153878ffcdb5ec06dfb83f8f1cccd8a06b87f59", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "babcfbed37fe5fa3be243feace9f3ca991b9dc0a540ca67e923206c246a3e227", "generation_seconds_batch": 23.929129087016918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1859106719493866, "target_z": -0.12353517355528801, "ce": 5.982900142669678, "genuineness": 4.7081804275512695, "identity": 0.5719485878944397, "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/emo_Pain/784bdf4713e2c0d3f82251f25153878ffcdb5ec06dfb83f8f1cccd8a06b87f59.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "78b7e96de18a6cbab4e232e36d8181e36fe436c8ed937c3f611cbdc7192ff781", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c5824495bb5f829efa30e81ec37079b3505e1af14df49e59b98092b054097753", "generation_seconds_batch": 22.915941952029243, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09969110786914825, "target_z": -0.27343750809604117, "ce": 6.443574905395508, "genuineness": 4.582659721374512, "identity": 0.571057140827179, "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/emo_Pain/78b7e96de18a6cbab4e232e36d8181e36fe436c8ed937c3f611cbdc7192ff781.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "78c90cf15bc7db82cb00f56fc52112c32bc025ee2a653d01db7428b77793d43e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "64bf3643c52a27ca7018214556f7a3e97bef6773bb60efebbf90f13597afc43a", "generation_seconds_batch": 23.013096615090035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6052130460739136, "target_z": 0.6054687993858512, "ce": 5.847874641418457, "genuineness": 4.7332844734191895, "identity": 0.5996526479721069, "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/emo_Pain/78c90cf15bc7db82cb00f56fc52112c32bc025ee2a653d01db7428b77793d43e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "78cba49ef5ee1565b36a411a2bf284e66b58fb72b26123735752dedd71e7dcc1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "304e43375e0b70853d0035d371b9be8616fc56712a555e22d6ac73ea3d9a7156", "generation_seconds_batch": 23.812703671050258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.060372740030288696, "target_z": -0.3417968786432185, "ce": 5.855817794799805, "genuineness": 4.419482707977295, "identity": 0.5774674415588379, "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/emo_Pain/78cba49ef5ee1565b36a411a2bf284e66b58fb72b26123735752dedd71e7dcc1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "7945d1946274a41673b7cca858bc97136969c4108d627749a56ef9260867a88a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3d25fd27ded5f744d6b490cf0396acb459c750936c98cd9bd5349a1e1645dd71", "generation_seconds_batch": 17.636291940929368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07048317790031433, "target_z": -0.3242187540480206, "ce": 6.117925643920898, "genuineness": 4.7332844734191895, "identity": 0.778306245803833, "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/emo_Pain/7945d1946274a41673b7cca858bc97136969c4108d627749a56ef9260867a88a.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "c7e6584e180fc29d4d41a7bb42a70b3cf45202a2dd59825be808400f84539803", "generation_seconds_batch": 17.1347317219479, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02442452311515808, "target_z": -0.40429686568955264, "ce": 6.213237762451172, "genuineness": 4.457139015197754, "identity": 0.7381144762039185, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de la en daarna naast het raam. Neem ze mee als je weggaat.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/794d3b9296e3c4be7c1e4182a2fa3cf095f7d593990763ce0f19c1f5e499a989.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "797e9bc1886eca3eb69e873238eedba9542f0f7e2995925b20438d2d2c27c7d6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ebb96eed2733404da2718aaaa90ab821b173904757fe92268f7e2cb557e046df", "generation_seconds_batch": 22.979316942975856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6007194519042969, "target_z": 0.5976561844220665, "ce": 6.078212261199951, "genuineness": 4.532451152801514, "identity": 0.48703503608703613, "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/emo_Pain/797e9bc1886eca3eb69e873238eedba9542f0f7e2995925b20438d2d2c27c7d6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "799123d256ff1036923373eef7e314a005deaae36f28e3e549703a7a1ce6621a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dc3f737b4bc1303c8093fd4e46499cdfe4aa8fadd9b31ea910c65ebabd15dcf7", "generation_seconds_batch": 22.13547286496032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017684221267700195, "target_z": -0.4160156326912391, "ce": 6.141753673553467, "genuineness": 4.444586753845215, "identity": 0.6196768283843994, "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/emo_Pain/799123d256ff1036923373eef7e314a005deaae36f28e3e549703a7a1ce6621a.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": "7a0eb1b91526f5de56351ab8576ce8cd25bb7bbc1c5f302f30d9af23bac6d29e", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "70e37ca8f9911ae06462bab1859dd52e9348503e4d62bc4b9bb00dc9b8d7b401", "generation_seconds_batch": 17.40354120999109, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44119924306869507, "target_z": 0.3203124983807918, "ce": 5.71682071685791, "genuineness": 4.293961524963379, "identity": 0.4147336483001709, "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/emo_Pain/7a0eb1b91526f5de56351ab8576ce8cd25bb7bbc1c5f302f30d9af23bac6d29e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "7a267cfcdbde21dc50260b295333ba1e262e97fe3d93b5415d7c4628ffd8640a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "65714718139884d2adfdf84fd1a3c6b8fe50b84fdee097bad28447156367d440", "generation_seconds_batch": 23.972587067983113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15024331212043762, "target_z": -0.1855468592127197, "ce": 6.284721851348877, "genuineness": 4.7834930419921875, "identity": 0.6662437915802002, "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/emo_Pain/7a267cfcdbde21dc50260b295333ba1e262e97fe3d93b5415d7c4628ffd8640a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "7a3eeceab96938c252c3e9bb67c5d4a845e35c1bcb929e485e83fb71f0290591", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "156bafa5c416b98b407887219e101948b9797b3e47c9a914e0db54a5380b8e39", "generation_seconds_batch": 27.929668414988555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14350301027297974, "target_z": -0.1972656262144062, "ce": 5.855817794799805, "genuineness": 4.632867813110352, "identity": 0.7194639444351196, "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/emo_Pain/7a3eeceab96938c252c3e9bb67c5d4a845e35c1bcb929e485e83fb71f0290591.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "7ab00463362a4a52e55a9d9004bf460045a707f6780dc2133dc6984f2a932b9b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "88deb5f1fbc77707c6b7a20640c9d5a6e29c3bc1e09f762aef66391a1bdb3490", "generation_seconds_batch": 17.747265412006527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5151524543762207, "target_z": 2.1875000129536657, "ce": 5.700935363769531, "genuineness": 4.005263805389404, "identity": 0.33704906702041626, "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/emo_Pain/7ab00463362a4a52e55a9d9004bf460045a707f6780dc2133dc6984f2a932b9b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "7b01efc463285d7a7ef82c1e8e82f56a26cce93e93961f95ea1d823e9f607d8c", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e63ca6e16d852480f85cec32b3b6c0910d18ac6df38c15cccfe186a374701dc4", "generation_seconds_batch": 18.01512525510043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.300607204437256, "genuineness": 4.607763767242432, "identity": 0.5776841640472412, "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/emo_Pain/7b01efc463285d7a7ef82c1e8e82f56a26cce93e93961f95ea1d823e9f607d8c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "7b335ef5bcc74478023e4b91fa5fcf5c1b77162b385797f95c5896a07ee92035", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "12df96bd3ae597860b03dfd409158dadaa3825e685645687021ab237c7443e1b", "generation_seconds_batch": 22.53379259898793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41873160004615784, "target_z": 0.28124999352316704, "ce": 5.442798614501953, "genuineness": 4.143336772918701, "identity": 0.45675814151763916, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/7b335ef5bcc74478023e4b91fa5fcf5c1b77162b385797f95c5896a07ee92035.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "7b74a7f08380b74130ef217541be30c2ef826634ec8c2f5d4b3b34e93d96492b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "22473f0cce11309cd6e6cb1adb2dbe334112d98bbefc61e37016ce05883fc18c", "generation_seconds_batch": 22.60858390200883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7220447063446045, "target_z": 0.8085936795644418, "ce": 5.6334228515625, "genuineness": 4.545003414154053, "identity": 0.4560551047325134, "hypothesis": "Die Tür ist noch offen. Ich hab dein Mantel auf den Stuhl gelegt. Sag mir, wenn du bereit bist.", "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/emo_Pain/7b74a7f08380b74130ef217541be30c2ef826634ec8c2f5d4b3b34e93d96492b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "7bca20d4bbf1a073ea945e90f53c6441e2252cdcc5935e0f8fef6541f70cee0d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2ec015cbe860d2466b62bf8e12d14f58a8e097878e63759e110c9590cacf0851", "generation_seconds_batch": 24.43903627502732, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 6.205295085906982, "genuineness": 4.507347106933594, "identity": 0.8790718913078308, "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/emo_Pain/7bca20d4bbf1a073ea945e90f53c6441e2252cdcc5935e0f8fef6541f70cee0d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "7bd30528b55f4e2d9e24e6333bc26dacf63ddabc661a2a6b3db620301d9fbac6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8da5061fe2782e1657fc751268dcc3816ba14c417df2a5a5f7faaef7ba926e2a", "generation_seconds_batch": 17.27680012001656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5804985761642456, "target_z": 0.5624999352316706, "ce": 6.173524379730225, "genuineness": 4.457139015197754, "identity": 0.6151471138000488, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/7bd30528b55f4e2d9e24e6333bc26dacf63ddabc661a2a6b3db620301d9fbac6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "7bdc7c9463df04bf0af04a4b52d72e4ae4a527e1b6f26d838dfa5d8e1ae3088b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0e775e625c82e5f985f49a5e755ce9c740e868579d9a452cadb94a9fd323a8ed", "generation_seconds_batch": 23.309764986042865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2620198130607605, "target_z": 0.00878906229759897, "ce": 5.744620323181152, "genuineness": 4.7834930419921875, "identity": 0.4224724769592285, "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/emo_Pain/7bdc7c9463df04bf0af04a4b52d72e4ae4a527e1b6f26d838dfa5d8e1ae3088b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "7c0e17580d41ea7cd34de657e842fa9ba0de956dfbb177880e558a1d8a5253e7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "90a6ef961ec13e033bd85b0761022aa6af8a03b417a977af8f0ab01172e327d3", "generation_seconds_batch": 23.929129087016918, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14799654483795166, "target_z": -0.18945311487994854, "ce": 5.974957466125488, "genuineness": 4.7458367347717285, "identity": 0.5805772542953491, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/7c0e17580d41ea7cd34de657e842fa9ba0de956dfbb177880e558a1d8a5253e7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "7c5b1947b98c32243e01cf000b400ff3a563ebc1d7b08b999e03ffd2badfbb57", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "12aea5342714af07226487a8a834e7169726bd9214b2184e3560bf14e4bc924a", "generation_seconds_batch": 21.69359755294863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4702171087265015, "target_z": 2.1093748996090893, "ce": 5.506340026855469, "genuineness": 4.406930446624756, "identity": 0.047594938427209854, "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/emo_Pain/7c5b1947b98c32243e01cf000b400ff3a563ebc1d7b08b999e03ffd2badfbb57.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "7c6e8627c6338e890ae44bb6f506656f9cec68bfa43ac2d79d592f6abab7d59f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7f51b29efa7afb1132b85fe4f45ec445207ef20dd4e54c1b433cbf60f68e40e8", "generation_seconds_batch": 18.003477772930637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20978254079818726, "target_z": -0.08203126538247824, "ce": 6.157639026641846, "genuineness": 4.557555198669434, "identity": 0.3482148349285126, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/7c6e8627c6338e890ae44bb6f506656f9cec68bfa43ac2d79d592f6abab7d59f.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": "7dae5db0fd6f667d32ef0547a76c993bbbb685afb8bc6446779371e5180371a2", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4f5362a8cf1968b1ec2b4e7cd3b53704b589c2c5ff41f0a5976c060e8bc4b5eb", "generation_seconds_batch": 23.666788005037233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3190314471721649, "target_z": 0.10791015088637272, "ce": 5.871703147888184, "genuineness": 4.545003414154053, "identity": 0.8296126127243042, "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/emo_Pain/7dae5db0fd6f667d32ef0547a76c993bbbb685afb8bc6446779371e5180371a2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "7de26a6c3a4284f9614ba623f3514b1e639b1ab2573cabb7afc71fd3da2665bc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "cfc97f2265c474495d0c22a967d882c22fc37759413b2118d41bcfbbe3c4e802", "generation_seconds_batch": 23.781702080043033, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3479585349559784, "target_z": 0.15820312135678147, "ce": 5.895530700683594, "genuineness": 4.7834930419921875, "identity": 0.5954275727272034, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/7de26a6c3a4284f9614ba623f3514b1e639b1ab2573cabb7afc71fd3da2665bc.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": "7e3752754c23d39f2f09cd99c33d6b311766465c78c1aaad4a1763daecbb87db", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "b1d2655da5a655a4926f293bf709a097a70edc1d03103435166cfc9f38bf1c54", "generation_seconds_batch": 18.72716961801052, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6119533181190491, "target_z": 0.6171875145728741, "ce": 6.141753673553467, "genuineness": 4.118232727050781, "identity": 0.17259997129440308, "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/emo_Pain/7e3752754c23d39f2f09cd99c33d6b311766465c78c1aaad4a1763daecbb87db.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "7e37b92de51ac07124a346bc89deaf2e2df73288584c1991e08322296a5d1b22", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0b769dd56ae93d982692b64e548736ccf396a454423b30f0b3d2f31aacadf310", "generation_seconds_batch": 17.56555771001149, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 6.268836498260498, "genuineness": 4.632867813110352, "identity": 0.7940405011177063, "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/emo_Pain/7e37b92de51ac07124a346bc89deaf2e2df73288584c1991e08322296a5d1b22.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "7e3b7fae11c1e878df486c76b28089c61f0da6a974335209a977acdc37e04bc3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "66fb1cab21e62180fcffbfd3b286f66d768dda5d0a52587301227465e30f6f8c", "generation_seconds_batch": 25.119574659969658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11204831302165985, "target_z": -0.2519531278336144, "ce": 6.252951145172119, "genuineness": 4.6705241203308105, "identity": 0.4909481704235077, "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/emo_Pain/7e3b7fae11c1e878df486c76b28089c61f0da6a974335209a977acdc37e04bc3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "7e5bafc669ff0d1b816ae8adac9362e279a3eef9baa0e12bc2e7fbbb943d4249", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b6711030db3e189856ef1122c8eaa69d4ebc90e4614a3dc192419d154ca3dbeb", "generation_seconds_batch": 17.636291940929368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0963209718465805, "target_z": -0.27929686568955264, "ce": 6.141753673553467, "genuineness": 4.68307638168335, "identity": 0.7822458744049072, "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/emo_Pain/7e5bafc669ff0d1b816ae8adac9362e279a3eef9baa0e12bc2e7fbbb943d4249.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "7ebde43c0704041bed6588a19d1a41e5da0de42844bd63225c7bf540d37e9139", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "25e9c99f72de95c232dd6343ea99734be38987615abd79736186910969bcb97d", "generation_seconds_batch": 23.7774008710403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6883432865142822, "target_z": 0.75, "ce": 5.879645347595215, "genuineness": 4.570107460021973, "identity": 0.6696628928184509, "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/emo_Pain/7ebde43c0704041bed6588a19d1a41e5da0de42844bd63225c7bf540d37e9139.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "6c7125ea73eb12fbe4ecaf9e4a6f857f9ca9f61f2606e51b6e2e851cdf4a2a70", "generation_seconds_batch": 17.76765338901896, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04689215123653412, "target_z": -0.3652343867392597, "ce": 6.348263263702393, "genuineness": 4.425758361816406, "identity": 0.322102427482605, "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/emo_Pain/7f6fce1a6f733719f45a363f28fee4e24b2f07a3396aa07b42392913d7c24239.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "7f8e8a8350cb3df2b2d507f566dba0a404c5552bb085bfdb9cf87456cc0271e0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 1.0, "cfg": 1.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f78a1992c9221471b903fd405f31bd782061d44d9fbab60d78955d91e4fea9aa", "generation_seconds_batch": 24.254576132982038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06823641061782837, "target_z": -0.3281250097152494, "ce": 6.086154937744141, "genuineness": 4.300237655639648, "identity": 0.3344768285751343, "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/emo_Pain/7f8e8a8350cb3df2b2d507f566dba0a404c5552bb085bfdb9cf87456cc0271e0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "7fabf1000eafcc338f3f1106a3bec00417855043ddcecef006a92f270aca311f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "213b3d312300cb0acfd948121776f45ab2437353c33404243d54050e9c2892e7", "generation_seconds_batch": 21.96024641301483, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20051464438438416, "target_z": -0.09814453762563242, "ce": 6.0305562019348145, "genuineness": 4.7081804275512695, "identity": 0.7740739583969116, "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/emo_Pain/7fabf1000eafcc338f3f1106a3bec00417855043ddcecef006a92f270aca311f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "7ff374863a546fdc1f59b3ac62966ec14705a448553df0c18fbb545f7e7dbab6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d96999b3dbb7ef3aace0e597643e93c254571664c64b34b9bd40bcfa0feae0b", "generation_seconds_batch": 16.694473252049647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6409711837768555, "target_z": 2.4062499158011716, "ce": 5.935244083404541, "genuineness": 4.432034492492676, "identity": 0.6861607432365417, "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/emo_Pain/7ff374863a546fdc1f59b3ac62966ec14705a448553df0c18fbb545f7e7dbab6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "8059425ebfb89ef637047770d326b944898493927bd4e293fe7aa5871aad17ff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2ac4600e5983c21c3e4409b15c2365168e62422338d6e0dbd739d3cae588dd42", "generation_seconds_batch": 16.113576994044706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1148567646741867, "target_z": -0.24707032120324426, "ce": 5.998785495758057, "genuineness": 4.632867813110352, "identity": 0.6878837943077087, "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/emo_Pain/8059425ebfb89ef637047770d326b944898493927bd4e293fe7aa5871aad17ff.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "805f60d102f9c78f96c82d5ab594ca482f80dd14e36d0b4a1e6fec60c3c60f59", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "862596534c4ad7d64d4dbbbaba33027a8a94931ad105da55d1f77fabfb050834", "generation_seconds_batch": 19.04886491491925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2732536196708679, "target_z": 0.028320288819079567, "ce": 6.292664527893066, "genuineness": 4.6203155517578125, "identity": 0.3901866674423218, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/805f60d102f9c78f96c82d5ab594ca482f80dd14e36d0b4a1e6fec60c3c60f59.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "808837cc9cefcd3f89fbd2baddea95d5ab424fdd42e62ba0c19a7f75c748121b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "40fae7b346a1946eb2b05e22a0bb1ecb63a87fd2e5de23ee7bc53e1e70779266", "generation_seconds_batch": 17.55001882102806, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28631293773651123, "target_z": 0.051025370739098866, "ce": 5.919358730316162, "genuineness": 4.68307638168335, "identity": 0.6864973306655884, "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/emo_Pain/808837cc9cefcd3f89fbd2baddea95d5ab424fdd42e62ba0c19a7f75c748121b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "808ff6d8308389e149313df977340ae441cbefb56e67a2098302dde25979a85b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d86a3362ebd86fedf8ad984b607c94d51994b0178f124a569c37121b94eace89", "generation_seconds_batch": 22.965865625068545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10193787515163422, "target_z": -0.2695312524288124, "ce": 6.09409761428833, "genuineness": 4.557555198669434, "identity": 0.4442784786224365, "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/emo_Pain/808ff6d8308389e149313df977340ae441cbefb56e67a2098302dde25979a85b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "80c10b39048cd1570e63151f08d08f4dde758b6091636c6eb685fcc0a7645b6b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2be77384982ccec4bbc36b6138f992dfea53f882d5ee5e9a8c62de9aa04914b2", "generation_seconds_batch": 21.492151062004268, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02779465913772583, "target_z": -0.39843750809604117, "ce": 5.927301406860352, "genuineness": 4.632867813110352, "identity": 0.7079089283943176, "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/emo_Pain/80c10b39048cd1570e63151f08d08f4dde758b6091636c6eb685fcc0a7645b6b.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "06f85752b7ea7020d2c4d25177d921e4d8ee98bc3c2d03f6b84b8452763e5ec9", "generation_seconds_batch": 16.675313247949816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.058125972747802734, "target_z": -0.34570313431044736, "ce": 6.109982967376709, "genuineness": 4.444586753845215, "identity": 0.7040157318115234, "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/emo_Pain/80c2597a6b8f43b5338a6b37d654868d679f46ded48514b42837ab18fba327ad.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "811bf9a5f533bba57b0f5abddc5905ac2ee66c8cd6de5ed6c780544a4a296149", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7e008239fbc743a82aa19836fa70cc711f2c0b2bb79e60c87157f1c4b0889f98", "generation_seconds_batch": 26.16772031004075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4569265842437744, "target_z": 0.34765623623673003, "ce": 5.911416053771973, "genuineness": 4.607763767242432, "identity": 0.7251989841461182, "hypothesis": "Ich habe die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir könnten 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/emo_Pain/811bf9a5f533bba57b0f5abddc5905ac2ee66c8cd6de5ed6c780544a4a296149.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "81588a36fc238c07d8f8f072e3499f34a5e2c6ccb48100a04fa99d9f5fb88661", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6c740a86f27c945698dc2121815f7e3827a521f4a2a87d6e4b2c64a9f66c0c90", "generation_seconds_batch": 16.411198070039973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7757771015167236, "target_z": 2.640625048576247, "ce": 5.6413655281066895, "genuineness": 4.356721878051758, "identity": 0.16466370224952698, "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/emo_Pain/81588a36fc238c07d8f8f072e3499f34a5e2c6ccb48100a04fa99d9f5fb88661.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "815a95fe263a4d7ae95ca179712eda8423f1065485646daf5b2acffadc23cc6c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a3c48b3b8dfc56e78e6b8147d6077517b278c9d78e66e67d9a1bbae6ef31172e", "generation_seconds_batch": 22.76200983207673, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08508715033531189, "target_z": -0.298828118118365, "ce": 6.23706579208374, "genuineness": 4.419482707977295, "identity": 0.4692254066467285, "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/emo_Pain/815a95fe263a4d7ae95ca179712eda8423f1065485646daf5b2acffadc23cc6c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "81651ae4552d2ec758757c9742432cb5832978cce5e68415662adfa2bf22738a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "54bdc9422fa11d9764f8a1d25a76477dfa0a3511ece69aec4378cbfad9ffedb0", "generation_seconds_batch": 22.23442521202378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44232261180877686, "target_z": 0.3222656003070744, "ce": 6.157639026641846, "genuineness": 4.68307638168335, "identity": 0.3851219415664673, "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/emo_Pain/81651ae4552d2ec758757c9742432cb5832978cce5e68415662adfa2bf22738a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "816d498809fa76c7f93231b21832c4d77979fc41966c74655141a93bff4ca0ec", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "00cf692f6acf0e5efcf686a874a72eafbecb3ced607b70dfec6110619af6f6f0", "generation_seconds_batch": 29.939873318071477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08845728635787964, "target_z": -0.29296876052485354, "ce": 6.109982967376709, "genuineness": 4.074300289154053, "identity": 0.2745944857597351, "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/emo_Pain/816d498809fa76c7f93231b21832c4d77979fc41966c74655141a93bff4ca0ec.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "827b7669f9baf3567c10011df7c04c65ee67f518f14c919dc7b9cdd29a48b6ee", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d7b36872319a5984d4ace314cb96d9a3bc6b616d56a1af941b7aae3ec07df371", "generation_seconds_batch": 16.812167576979846, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8411232233047485, "target_z": 1.0156249708542517, "ce": 5.871703147888184, "genuineness": 4.369274139404297, "identity": 0.606550931930542, "hypothesis": "De deur staat nog open. Ik heb je 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/emo_Pain/827b7669f9baf3567c10011df7c04c65ee67f518f14c919dc7b9cdd29a48b6ee.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "82826503b11c139e273f242f788434d4292c3444f0a3b849dda5a69c662059b9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c465ccf99096bbf8bd35307a9e9c5497458be7a79e4c2d1217f681bac01f6ab6", "generation_seconds_batch": 21.487092609051615, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11766523122787476, "target_z": -0.24218748866554235, "ce": 6.324435234069824, "genuineness": 4.720732688903809, "identity": 0.43497368693351746, "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/emo_Pain/82826503b11c139e273f242f788434d4292c3444f0a3b849dda5a69c662059b9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "83199f68fb8d743cd0a8b2967783770b25de20366e8e1dbfd0487d84be012ba6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "60e0839fa4869b6f96ec948839f4343c0ffad1c528baf6e4e13e62d94c2d6645", "generation_seconds_batch": 23.084421940962784, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13788610696792603, "target_z": -0.20703123947514648, "ce": 6.125868320465088, "genuineness": 5.034534454345703, "identity": 0.6118863224983215, "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/emo_Pain/83199f68fb8d743cd0a8b2967783770b25de20366e8e1dbfd0487d84be012ba6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "83246241a7f468175d15717508846001d457c780521894b1598c57cce7e698d3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2f9851cd146ffae7d8350ef1484e81708c1e1ce70e27b4e563337ce1160e539f", "generation_seconds_batch": 17.55427448498085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41199129819869995, "target_z": 0.2695312265214806, "ce": 5.935244083404541, "genuineness": 4.507347106933594, "identity": 0.7496528625488281, "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/emo_Pain/83246241a7f468175d15717508846001d457c780521894b1598c57cce7e698d3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "836964b42e9a87d587cd92e9b84d55d62feb1e71620f1355c239af636e4cd384", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c3b356b4c2bfe2c408a68f5da65370cfe755f8030ec253a6ac0f3473b0041a31", "generation_seconds_batch": 22.651471155928448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1266522854566574, "target_z": -0.22656249190395883, "ce": 6.41180419921875, "genuineness": 4.632867813110352, "identity": 0.5122904181480408, "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/emo_Pain/836964b42e9a87d587cd92e9b84d55d62feb1e71620f1355c239af636e4cd384.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "83a5069ae1a45b3e5143287ddb46ede07a38fd0ad640e5243e41a402c2431b58", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9b0623898c886457a8fb9575a9f6fa7255b630f1138b5f73c690fb20837cfb1b", "generation_seconds_batch": 22.86629430402536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1871248483657837, "target_z": 1.6171874109435471, "ce": 5.677107334136963, "genuineness": 4.331617832183838, "identity": 0.5772424936294556, "hypothesis": "Ich habe ihr keine Schachtel gefunden. Sie stand neben dem alten Pfände. Wir können gemeinsam hineinschauen.", "wer": 0.1875, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/83a5069ae1a45b3e5143287ddb46ede07a38fd0ad640e5243e41a402c2431b58.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "83cf2c588ef534d394abf4b11391ce06c42b41a6041ff202526b18f531a6a690", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "24bca02f76828df6795673c6aa4da877977dac47d3e2cfd456c92359d0807d43", "generation_seconds_batch": 23.836500382982194, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.1020402908325195, "genuineness": 4.607763767242432, "identity": 0.7166029810905457, "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/emo_Pain/83cf2c588ef534d394abf4b11391ce06c42b41a6041ff202526b18f531a6a690.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "83dad4ac3573fa7cb1eda9958c6282d3656def21810f6f3b3bf60a286fb00ff5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7cbd49d6a4b09d61aa9b8cccae550dc3aed9da0a8678c2eafa5b5b8b19693a49", "generation_seconds_batch": 25.578460563905537, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.24643288552761078, "target_z": -0.018310550771219816, "ce": 5.748591423034668, "genuineness": 4.394378662109375, "identity": 0.36797279119491577, "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/emo_Pain/83dad4ac3573fa7cb1eda9958c6282d3656def21810f6f3b3bf60a286fb00ff5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "847715091156e7c6a09a2d3df64351189033866d093de6a19d0fb5e3a03252d8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "96b47f5e1034d87db5656ee12e0c1d7479e9d3cef3c55450af3147d019b37c7a", "generation_seconds_batch": 24.454157738015056, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9354872703552246, "target_z": 1.1796873979898812, "ce": 5.5659098625183105, "genuineness": 4.419482707977295, "identity": 0.70548415184021, "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/emo_Pain/847715091156e7c6a09a2d3df64351189033866d093de6a19d0fb5e3a03252d8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "84995605050563fcca14b3b76f5820611e23ff3e866d03d4285f4b688b52e77e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6e020ed18f383dd3ab6c272ce22dcd2407cd5470b15929f6834fd2e7bd9be9a1", "generation_seconds_batch": 22.15373884292785, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04239863157272339, "target_z": -0.3730468721663856, "ce": 5.903473377227783, "genuineness": 4.507347106933594, "identity": 0.4519420862197876, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar, wanneer je klaar bent, ik.", "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/emo_Pain/84995605050563fcca14b3b76f5820611e23ff3e866d03d4285f4b688b52e77e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "84e186e0815605f8000d591cf9bd119b53b8c6a1af4a1526868838e40a8f6e84", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c44cab02f5cfe7ccd01ef5decfbe290731c010cca39abc7161f599bdfaba26a8", "generation_seconds_batch": 26.808179888990708, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1755194067955017, "target_z": -0.14160154772472486, "ce": 5.967014789581299, "genuineness": 4.984326362609863, "identity": 0.6773298978805542, "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/emo_Pain/84e186e0815605f8000d591cf9bd119b53b8c6a1af4a1526868838e40a8f6e84.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": "852f1d8e8d70b7c6b279c5b44c6a4fcd1cbc3f9224b5016ab161bdeafc58cc8a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Ik wacht aan de andere kant.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Natural clear speech.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Natural clear speech.) [3.2 seconds duration] Ik wacht aan de andere kant.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4bfdbd4fd37b7fff39669bac2219777d969bec2e3a32e673a8d79c6909369b85", "generation_seconds_batch": 17.38699952198658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5939792394638062, "target_z": 0.5859375728643705, "ce": 6.038498878479004, "genuineness": 4.256305694580078, "identity": 0.3977968990802765, "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/emo_Pain/852f1d8e8d70b7c6b279c5b44c6a4fcd1cbc3f9224b5016ab161bdeafc58cc8a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "85f662f68602b77c86d9ca263bb1b13e48cad71d423f6cd2f84e136749b829e7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9b3c7de9be059c8a627e1c658544e4b02516970c0a69ee342596558f95c58a67", "generation_seconds_batch": 25.651551496004686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11092492938041687, "target_z": -0.2539062556672288, "ce": 6.268836498260498, "genuineness": 4.494794845581055, "identity": 0.36951297521591187, "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/emo_Pain/85f662f68602b77c86d9ca263bb1b13e48cad71d423f6cd2f84e136749b829e7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "861a7a1f81aaeeeea60fa69542f3667cafaaec35e745027689249c93be57d2fb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "744609419e1cfc0d6b3e7206bf196de53e007db3ac9320e9e57340aa7f2b213d", "generation_seconds_batch": 16.573090917081572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5939792394638062, "target_z": 0.5859375728643705, "ce": 5.95112943649292, "genuineness": 4.770940780639648, "identity": 0.7389602065086365, "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/emo_Pain/861a7a1f81aaeeeea60fa69542f3667cafaaec35e745027689249c93be57d2fb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "863ab3b18c39c0075f8e65530bab69e240b03a965451a7a420a9be34cf90e581", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "53a0e7d644db5b48dc1425dc8c20771ef098cf2621a836e9e939a0316b22057d", "generation_seconds_batch": 23.458835119963624, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2994630336761475, "target_z": 1.8124998834170072, "ce": 5.982900142669678, "genuineness": 4.256305694580078, "identity": 0.5308207273483276, "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/emo_Pain/863ab3b18c39c0075f8e65530bab69e240b03a965451a7a420a9be34cf90e581.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "864940e37ce1c22a686a232bb4d564f050b2c1b421dfda27aadb60c68d2e92fc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6add45ed9cb1777891d2914acf723260b758f4cdc0457adb0a20ae7d3bdda15e", "generation_seconds_batch": 23.933782907901332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14911991357803345, "target_z": -0.1875000129536659, "ce": 5.824047088623047, "genuineness": 4.6956281661987305, "identity": 0.5890264511108398, "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/emo_Pain/864940e37ce1c22a686a232bb4d564f050b2c1b421dfda27aadb60c68d2e92fc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "86866b5e31bddafa9ae0935d513a781d1905560d41ed734bb55df0de8c7b136c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "369f28bc607864bb39cc20dd528792eaf7a7f4df582f44e87be5b3ea143e2c7b", "generation_seconds_batch": 27.67689431400504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04239863157272339, "target_z": -0.3730468721663856, "ce": 6.38003396987915, "genuineness": 4.595211505889893, "identity": 0.7537825107574463, "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/emo_Pain/86866b5e31bddafa9ae0935d513a781d1905560d41ed734bb55df0de8c7b136c.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": "86db7ed15c033bb0ff049fec55bf2a64c4a8539ab63ca3eaa62f13e728bcdbfc", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "66674bbcebd2f7c128de5a0d81f8b5ee813827b17b1e5e96d226c957464df0e7", "generation_seconds_batch": 21.540408671018668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061306118965149, "target_z": 1.3984375080960412, "ce": 5.538110733032227, "genuineness": 4.231201171875, "identity": 0.53337562084198, "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/emo_Pain/86db7ed15c033bb0ff049fec55bf2a64c4a8539ab63ca3eaa62f13e728bcdbfc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "87295f9ca1b8a212e3b4230e5350f6bdd61ca4d9ef36506938d2f42c21751565", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "de0e4531033bde5d8723c77709355b46a497848a4f5ab6ce932ebd96e11259ed", "generation_seconds_batch": 23.381369772017933, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.582745373249054, "target_z": 0.566406242713563, "ce": 5.6532793045043945, "genuineness": 4.7081804275512695, "identity": 0.7099503874778748, "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/emo_Pain/87295f9ca1b8a212e3b4230e5350f6bdd61ca4d9ef36506938d2f42c21751565.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "87a06f4f976f03771ded3c488cd4a2a1c11d7978f322727104c74ac529d2f9d7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4c4359564a9a348b906370e4249c928a39674487667ca4dbef125ba6758d4c55", "generation_seconds_batch": 22.503070921986364, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08059361577033997, "target_z": -0.30664062945282267, "ce": 6.157639026641846, "genuineness": 4.7332844734191895, "identity": 0.5198110342025757, "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/emo_Pain/87a06f4f976f03771ded3c488cd4a2a1c11d7978f322727104c74ac529d2f9d7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "87aa5ad7630382190e9df54945ea4dfa3164071150615e04dfa41a2a597f73ca", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7175705418263707aa4347e92038a22b8154f3fa80efa3338a1103dfc9f3562e", "generation_seconds_batch": 22.416539905942045, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31425708532333374, "target_z": 0.09960938997767617, "ce": 5.6413655281066895, "genuineness": 4.469690799713135, "identity": 0.6247364282608032, "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/emo_Pain/87aa5ad7630382190e9df54945ea4dfa3164071150615e04dfa41a2a597f73ca.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "87fb120a9af561acf90398280665a0b0a8ffb56192ddb5ceb3b4c4e66aadd043", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "98d9f32946b5be32ae025416dc3e3bb1fc1942a11877abef617b4bf0235fa0d0", "generation_seconds_batch": 23.633160769939423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 5.911416053771973, "genuineness": 4.482243061065674, "identity": 0.3914235234260559, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/87fb120a9af561acf90398280665a0b0a8ffb56192ddb5ceb3b4c4e66aadd043.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "881ad2088a9acdb968a0a0b9a75d1de051d0196a0b9ea3ed4a007bad6d41ed89", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "68ad1f4cd278890637c92989c9e720a624aeff0ddb346c5c3cc34949d17cf6e0", "generation_seconds_batch": 24.373335314099677, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05138568580150604, "target_z": -0.3574218754048021, "ce": 5.816103935241699, "genuineness": 4.162164688110352, "identity": 0.5722455978393555, "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/emo_Pain/881ad2088a9acdb968a0a0b9a75d1de051d0196a0b9ea3ed4a007bad6d41ed89.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "89203644a5bccb5f7b5938454eb3ca851af8956db437023d4f844c2a28d38252", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8a111d13176176f1ecb4e01d6c6934a7ed545b77122311e9bdcd7b58e57a04a2", "generation_seconds_batch": 25.651551496004686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.24500846862793, "genuineness": 4.362998008728027, "identity": 0.5160355567932129, "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/emo_Pain/89203644a5bccb5f7b5938454eb3ca851af8956db437023d4f844c2a28d38252.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "89aea54c09b71cccfcb242e318ba1bbde0938db73b31012659fed233c9f29c27", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "39f0c00c3bc8d971c9243f4e9c4f8151bd41b429fcc57130fbb1a432c706467e", "generation_seconds_batch": 22.114062309032306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13451595604419708, "target_z": -0.2128906229759897, "ce": 6.41180419921875, "genuineness": 4.770940780639648, "identity": 0.5373842716217041, "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/emo_Pain/89aea54c09b71cccfcb242e318ba1bbde0938db73b31012659fed233c9f29c27.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "89e28dd9f3b9b785213b4d119615e2b8439efc5e0f0f5d010bdc43cfb86f399f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a86ef9cff5775c5f543c76bd7e85852247edc5488416de8b830e0715ab6b365c", "generation_seconds_batch": 23.16149537707679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5331264734268188, "target_z": 2.2187498510328423, "ce": 5.319687366485596, "genuineness": 4.143336772918701, "identity": 0.47739970684051514, "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/emo_Pain/89e28dd9f3b9b785213b4d119615e2b8439efc5e0f0f5d010bdc43cfb86f399f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "8a43d0aac5b693402abd6a7947991c07871acb53d43239dd043d5ede41210370", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b01f5bdd66f8d8f5bfc8b4df895ca04c3e149adf5b9df9204e44bfa95c2a80fb", "generation_seconds_batch": 24.8094939639559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1017478704452515, "target_z": 1.4687500064768328, "ce": 5.498397350311279, "genuineness": 4.231201171875, "identity": 0.43435531854629517, "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/emo_Pain/8a43d0aac5b693402abd6a7947991c07871acb53d43239dd043d5ede41210370.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "8a8d23096555708aab29a7442a75733ac001988c823506ac8279b1fca2b050e2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bf9924504da642b7e967e0a075d35c29a3ed99c1c1c55bcdc69c2d3f49c694df", "generation_seconds_batch": 23.144475175999105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13620102405548096, "target_z": -0.20996094417923397, "ce": 5.998785495758057, "genuineness": 4.7081804275512695, "identity": 0.7100206613540649, "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/emo_Pain/8a8d23096555708aab29a7442a75733ac001988c823506ac8279b1fca2b050e2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "8b4ff6150ee480da58a54a51af073289d306c954fbe1da272cbc83901f108c9a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a5180cf4f41f135281b226edc398937ce231e4a74fd66fd891dde5bd1068f2c1", "generation_seconds_batch": 19.901732868980616, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5151524543762207, "target_z": 2.1875000129536657, "ce": 5.728734970092773, "genuineness": 4.174716949462891, "identity": 0.15174943208694458, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude ramen. 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/emo_Pain/8b4ff6150ee480da58a54a51af073289d306c954fbe1da272cbc83901f108c9a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "8bb5e93ab07bd863380f0c7152a5ce9190353d5b4ab9ff9a73fa0a5c9d60875f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a5f4bbd787a5789f1f2c70fc11ea38ced545007a132264d9636cd5da761387dd", "generation_seconds_batch": 22.140574865974486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.054755836725234985, "target_z": -0.35156249190395883, "ce": 6.0305562019348145, "genuineness": 4.394378662109375, "identity": 0.7517567873001099, "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/emo_Pain/8bb5e93ab07bd863380f0c7152a5ce9190353d5b4ab9ff9a73fa0a5c9d60875f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "8c6738743140a50bd799e89166c0e1d4d7985e66e31b8ecbbca6d6b68824e7c0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cf91abef45446d6404cd79606fc7a1d741436e5725bbe78b7f3e000f77e2f0bc", "generation_seconds_batch": 23.624915573047474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.031164810061454773, "target_z": -0.3925781245951979, "ce": 5.998785495758057, "genuineness": 4.15588903427124, "identity": 0.47052693367004395, "hypothesis": "Ik heb het kleine, het kleine doosje gevonden. Het stond naast het oude raam. We kunnen er 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/emo_Pain/8c6738743140a50bd799e89166c0e1d4d7985e66e31b8ecbbca6d6b68824e7c0.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": "8ccc22bb67a5add751f0b6e49e9c3f125d9ca3b9054fa0d581faea4452b38f1d", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We zetten het na de lunch terug.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Natural clear speech.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Natural clear speech.) [3.2 seconds duration] We zetten het na de lunch terug.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d94b2739685e4792cd1f3393380be64da0e411d2d076c50885ceb35cf55c6bc5", "generation_seconds_batch": 18.38001204095781, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 5.990842819213867, "genuineness": 4.406930446624756, "identity": 0.5834938287734985, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/8ccc22bb67a5add751f0b6e49e9c3f125d9ca3b9054fa0d581faea4452b38f1d.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": "8cd0571618ca6fbc26fd059d241aae0ac6140627d88097ad173d83d360126c68", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "b4647da7b579301def64a99f76a389719474b1585f47bc9c36cbc339c0ecba4b", "generation_seconds_batch": 19.075753710931167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 6.054384231567383, "genuineness": 4.7081804275512695, "identity": 0.6920934915542603, "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/emo_Pain/8cd0571618ca6fbc26fd059d241aae0ac6140627d88097ad173d83d360126c68.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "8ce6a6e11f28de0de60d23e2ba4aea1112078f1b2a2b84034940bc62b3f1c770", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c31b966bf1303500b9ecb91c89e0898353f4ee71d433461d488f5f50c23cd391", "generation_seconds_batch": 22.558167097973637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5018618702888489, "target_z": 0.4257812459519794, "ce": 5.784333229064941, "genuineness": 4.3504462242126465, "identity": 0.6047203540802002, "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/emo_Pain/8ce6a6e11f28de0de60d23e2ba4aea1112078f1b2a2b84034940bc62b3f1c770.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "8cfbca194ade0f170ab46528938d0b7a8eafd695a23edf5f0b1243a01bf4793b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1f38a4d3cefc4b1fa8159f9016276f50f480f28cd9e1a551afa3d3cc09d257a3", "generation_seconds_batch": 16.795798577019013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2478371113538742, "target_z": -0.01586914745603474, "ce": 5.784333229064941, "genuineness": 4.519899368286133, "identity": 0.597687304019928, "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/emo_Pain/8cfbca194ade0f170ab46528938d0b7a8eafd695a23edf5f0b1243a01bf4793b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "8d4760a61e025a906e6e7a1bf18aea32ba28cb124d0a828e48c8967c1022c171", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2afbfc70509511e27b99994ad0e943b9a670929b570e6891448a6007b6d01e9b", "generation_seconds_batch": 22.95724389597308, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4389524757862091, "target_z": 0.31640624271356294, "ce": 5.681078910827637, "genuineness": 4.444586753845215, "identity": 0.633741557598114, "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/emo_Pain/8d4760a61e025a906e6e7a1bf18aea32ba28cb124d0a828e48c8967c1022c171.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": "8d7e497b16123e295f38959e8f3a025d13dce1af7fc4f65a6213049ab647a92c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fab74729769ea43e6ecbf346e96ea7a13ff37ac16feb1bbb7cb04cbcc17f196f", "generation_seconds_batch": 18.676699334057048, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4996151030063629, "target_z": 0.4218749902847506, "ce": 5.80021858215332, "genuineness": 4.532451152801514, "identity": -0.003243023529648781, "hypothesis": "De sleutels liggen op de keukentafel. Ik keek in de lo 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/emo_Pain/8d7e497b16123e295f38959e8f3a025d13dce1af7fc4f65a6213049ab647a92c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "8da7f126557b7d07effdac454b46b731e347a16a4e47bb557f14fd43534b7259", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "5895299bda53f0e7297663cd14e3d1b8a71c966951a628c523e4e4559cf8b817", "generation_seconds_batch": 17.76864701299928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.047825574874878, "target_z": 1.3750000777219953, "ce": 5.895530700683594, "genuineness": 4.457139015197754, "identity": 0.4583446979522705, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de la vieille fenêtre. Nous pouvons ch regarder ensemble ce qu'elle contient.", "wer": 0.05263157894736842, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/8da7f126557b7d07effdac454b46b731e347a16a4e47bb557f14fd43534b7259.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "8dbb14fdd50f5519881fb0e4d1c84ce6e04196957b8741c8b0f85769913c4447", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "61c3d75d31386e0573fd500af3ff70347e1db9bc07ffd6585cae7831f2b2b6d0", "generation_seconds_batch": 23.849411083036102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9309937953948975, "target_z": 1.1718749902847505, "ce": 6.070269584655762, "genuineness": 4.796044826507568, "identity": 0.6468617916107178, "hypothesis": "La porte est encore ouverte. J'ai 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/manual-comparison-v1/audio/emo_Pain/8dbb14fdd50f5519881fb0e4d1c84ce6e04196957b8741c8b0f85769913c4447.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "8e16bd9ef1fcc0ebf46b67bb3110c3fdbe1157c37ad232668e450eb646c40830", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "91718431fa647ef3bd963729e0f813bd5ef86691ad5e16dc3e9babfc83506099", "generation_seconds_batch": 22.221234102966264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13788610696792603, "target_z": -0.20703123947514648, "ce": 6.117925643920898, "genuineness": 4.808597087860107, "identity": 0.7623655200004578, "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/emo_Pain/8e16bd9ef1fcc0ebf46b67bb3110c3fdbe1157c37ad232668e450eb646c40830.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "8e1f0b2e1e97fac1b9b6ea92ccbebe125b59f0812b12f78d907fed3c61649202", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2173b37bedae2a11ec98a663270d2bd1fe9a9364a008f045036436d2b687d24e", "generation_seconds_batch": 16.336989733972587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2614581286907196, "target_z": 0.007812511334457643, "ce": 5.998785495758057, "genuineness": 4.883909225463867, "identity": 0.71018385887146, "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/emo_Pain/8e1f0b2e1e97fac1b9b6ea92ccbebe125b59f0812b12f78d907fed3c61649202.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": "8e2e647f4cd9708d426895d6657cbe778a817299db5c773dd6e5585a9e5c1f3f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "81f4fa321a2917d087bad17acf1be22e1b8132d6ae29dc74aaead4e4d67f46c5", "generation_seconds_batch": 18.02509222610388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04352201521396637, "target_z": -0.3710937443327712, "ce": 5.792276382446289, "genuineness": 4.319066047668457, "identity": 0.1610577404499054, "hypothesis": "Die sleutels liggen op de keukentafel. Ik keek 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/emo_Pain/8e2e647f4cd9708d426895d6657cbe778a817299db5c773dd6e5585a9e5c1f3f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "8e71049fcbbe7bc8a66d2a821329c9cee60d7fe7d249562ac9fb28ee308e3003", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "5f943b864c393eaba7fdffdc031a529cb19f7b0ea0a32a3cf3f35274d212aa21", "generation_seconds_batch": 16.6087052320363, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.078346848487854, "target_z": -0.31054688512005146, "ce": 5.998785495758057, "genuineness": 3.9801597595214844, "identity": 0.20439347624778748, "hypothesis": "The door is still open. I've put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "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/emo_Pain/8e71049fcbbe7bc8a66d2a821329c9cee60d7fe7d249562ac9fb28ee308e3003.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": "8f1fdc1ad8a8e0e97a5771ea8084053da0da44b147487286cf8e6161b58aca75", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7d0c06e145a49849613feb76db8ab321d6992116d313080d143245745cc7d8b3", "generation_seconds_batch": 22.86036586901173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.43558233976364136, "target_z": 0.31054688512005146, "ce": 6.149696350097656, "genuineness": 4.042920112609863, "identity": 0.7196194529533386, "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/emo_Pain/8f1fdc1ad8a8e0e97a5771ea8084053da0da44b147487286cf8e6161b58aca75.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "8f483b271dd376807fb69b5a81fbb793b6ab64889b8adc921634de2c65dafe89", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "faeecf2e556172d7017a239990b616c6adefc325e8f2f8ba0b39c36b22bf794c", "generation_seconds_batch": 23.65452500700485, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1646572351455688, "target_z": 1.5781249579005858, "ce": 5.5817952156066895, "genuineness": 3.9801597595214844, "identity": 0.21319639682769775, "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/emo_Pain/8f483b271dd376807fb69b5a81fbb793b6ab64889b8adc921634de2c65dafe89.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "8f7150971dd1c5361c5c7a26d0d0090d9b9061205bcc57241360c728e5d6fc84", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6929416c97c3df3d9e3ebb64c5adc7fef5da4aa742ac77449155dc8c98195f1e", "generation_seconds_batch": 18.232507715001702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030041426420211792, "target_z": -0.3945312524288124, "ce": 6.157639026641846, "genuineness": 4.65797233581543, "identity": 0.6625058054924011, "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/emo_Pain/8f7150971dd1c5361c5c7a26d0d0090d9b9061205bcc57241360c728e5d6fc84.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "8f834e59f9b60f99da8b79417e7728fc770778eb6ce8a1c9f8235fe4a034db73", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c1aa43559e1248868e0a005cc69da5bfa18b10119ca5fc73c0ed34ebde0c68f8", "generation_seconds_batch": 23.39202202996239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7445123791694641, "target_z": 0.84765623623673, "ce": 6.1894097328186035, "genuineness": 4.306513786315918, "identity": 0.670089840888977, "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/emo_Pain/8f834e59f9b60f99da8b79417e7728fc770778eb6ce8a1c9f8235fe4a034db73.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "8fb63bccbeb2f9b01a8b05f32a1ccf5655d57068207ed4c6f115f7d3e6018768", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6efbad97b69e0b230b3839153acca74318f8d8f63f47c1a3501b2295d2adf9f0", "generation_seconds_batch": 24.62328832608182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08845728635787964, "target_z": -0.29296876052485354, "ce": 6.268836498260498, "genuineness": 4.959221839904785, "identity": 0.7162303924560547, "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/emo_Pain/8fb63bccbeb2f9b01a8b05f32a1ccf5655d57068207ed4c6f115f7d3e6018768.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "8fc14c50e80adfbe3a6aa7519af905acc5f235d255228f9abbf315a10af112ac", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "d622ada1d0b8c9c32dbed183a70e55cb684783f38bba20d64c33efec7b02bfdb", "generation_seconds_batch": 22.476641788030975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5061653852462769, "target_z": 2.1718749902847505, "ce": 5.673136234283447, "genuineness": 4.413206577301025, "identity": 0.6528090238571167, "hypothesis": "La porte est encore ouverte. J'ai posé ton manteau sur la chaise. Dis-moi quand tu es prêt. C'est Dracique quand tu es prêt.", "wer": 0.3333333333333333, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/8fc14c50e80adfbe3a6aa7519af905acc5f235d255228f9abbf315a10af112ac.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": "905ea84e4449386a19ad640dfa9d39655e446e2a68f267d646d252f53690e578", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Llévatelas cuando te vayas.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Natural clear speech.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Natural clear speech.) [3.2 seconds duration] Llévatelas cuando te vayas.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8d41a4911c33c9cfedebccc4bd9e43c4f47b4c41b9172387bc0bfcff14a29c4a", "generation_seconds_batch": 17.663815699983388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32549089193344116, "target_z": 0.11914061649915676, "ce": 5.665193557739258, "genuineness": 4.469690799713135, "identity": 0.6963343620300293, "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/emo_Pain/905ea84e4449386a19ad640dfa9d39655e446e2a68f267d646d252f53690e578.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "908a50f5b609b3ec6c025e6750be5ce01fc8f48157700110b747832af4525aea", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "f84ada0cc9660cddb7fc7627517e9db4d4f957d910bda18495122848a6908d87", "generation_seconds_batch": 22.603523959987797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8680844306945801, "target_z": 1.0625000388609975, "ce": 5.657250881195068, "genuineness": 4.218649387359619, "identity": 0.17753760516643524, "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/emo_Pain/908a50f5b609b3ec6c025e6750be5ce01fc8f48157700110b747832af4525aea.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "90c7666b9d79787365ad2aebb90621b8dad4ac9b1334d97f3dde8c97569140fb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "67598ec0d4b488a6396d161c71cf316eb3a3c795195c00ce97d9b4b9fac13ad1", "generation_seconds_batch": 22.142778928973712, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10643140971660614, "target_z": -0.2617187410943547, "ce": 6.308549880981445, "genuineness": 4.645420074462891, "identity": 0.5214720964431763, "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/emo_Pain/90c7666b9d79787365ad2aebb90621b8dad4ac9b1334d97f3dde8c97569140fb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "9305ec3ca63cfb8ccb6e005dde8cb7e7f1618e3b5df9d09b46f7cda104641862", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2e97db20211275f08c0d49303a081e1ec3704ec02d5082b7b61f78b69ae48d4b", "generation_seconds_batch": 18.53094234294258, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39401721954345703, "target_z": 0.23828128481297706, "ce": 5.736677646636963, "genuineness": 4.7332844734191895, "identity": 0.49953216314315796, "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/emo_Pain/9305ec3ca63cfb8ccb6e005dde8cb7e7f1618e3b5df9d09b46f7cda104641862.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "93243efbf8bcc9bd37b5490ead4c80a1d897864d06998f193f4ee3ac5cb717cf", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "ca78cae08092750463c1c76adca0b50d4a0892324380f1ce4e19734f705e6cb7", "generation_seconds_batch": 22.182554475031793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5467971563339233, "target_z": 0.5039062556672288, "ce": 6.070269584655762, "genuineness": 4.394378662109375, "identity": 0.7096048593521118, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/93243efbf8bcc9bd37b5490ead4c80a1d897864d06998f193f4ee3ac5cb717cf.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "936e3fff372d2a5073e5d3e7f9bc83740dcdc9f7ab7661171830da8f3a63fd9d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0c66359a58bf1f8112d0f24a7916d0e171457b3a7f78137df4f6ecb0db4a458b", "generation_seconds_batch": 24.38280466804281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23126722872257233, "target_z": -0.044677737664016726, "ce": 5.974957466125488, "genuineness": 4.632867813110352, "identity": 0.7657894492149353, "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/emo_Pain/936e3fff372d2a5073e5d3e7f9bc83740dcdc9f7ab7661171830da8f3a63fd9d.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": "93b0bc35cd54fa42e3eb8cee89e466d857969b77ee68b3de468837dbf1a61701", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "cc8960ec420330630d39ca598cd0e2c88b2fab51d1ed526ded354d4dc734837a", "generation_seconds_batch": 17.414270774112083, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10530802607536316, "target_z": -0.2636718689279691, "ce": 5.418970584869385, "genuineness": 4.444586753845215, "identity": 0.5418343544006348, "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": 0.03999999999999915, "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/emo_Pain/93b0bc35cd54fa42e3eb8cee89e466d857969b77ee68b3de468837dbf1a61701.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "93e430cc28fb357485b567c7c84df6597f53b927ffac6a6d35c935a0c43f964f", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b37b0a74c6c304da3f9d33ced3b200c5d6a587b6538a2ab7b4dd09ed99e14505", "generation_seconds_batch": 16.74245912209153, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.427689552307129, "genuineness": 4.7332844734191895, "identity": 0.479486882686615, "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/emo_Pain/93e430cc28fb357485b567c7c84df6597f53b927ffac6a6d35c935a0c43f964f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "944d751174a8784ced609bcb4b26fc43e795c6db6c1edca119a4b5d514e0cf79", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "05d900ce46b7cec87f64b7795ed273eac79587e6db2e97366f87dc597ba46460", "generation_seconds_batch": 17.657214643084444, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0356583446264267, "target_z": -0.3847656132607403, "ce": 5.998785495758057, "genuineness": 4.356721878051758, "identity": 0.5934679508209229, "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/emo_Pain/944d751174a8784ced609bcb4b26fc43e795c6db6c1edca119a4b5d514e0cf79.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "9490509df724700cb21b7a26897e8a165b26ec33d2b18d79f33576a502697da1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "794dfd229881453628f5fc62d2e393c9d6eb37e58816e5b08b3b7b5bd6c956bc", "generation_seconds_batch": 24.3197509549791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2968446612358093, "target_z": 0.06933594741765044, "ce": 6.221180438995361, "genuineness": 4.494794845581055, "identity": 0.6474391222000122, "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/emo_Pain/9490509df724700cb21b7a26897e8a165b26ec33d2b18d79f33576a502697da1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "94924f7a98ad3de99ae2378334ba41a5f93055ef2342b2fd22aa2cbdc2d618c5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5ad9e6b0f61c71ec78eb2b65bd8cedf345201575cf3a983f063e3e1966d72c47", "generation_seconds_batch": 28.277724315063097, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 5.935244083404541, "genuineness": 4.770940780639648, "identity": 0.5721396207809448, "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/emo_Pain/94924f7a98ad3de99ae2378334ba41a5f93055ef2342b2fd22aa2cbdc2d618c5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "94bb1c212cf7088b3ad33acd65c085ddce65db00a8508347d8aa3bb1dbac26b5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 0.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "4e50407953bb2a077aacddcfa45ddc7c674485ad28ac3db6644b4ce58cb357d3", "generation_seconds_batch": 18.43492199399043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5151524543762207, "target_z": 2.1875000129536657, "ce": 5.039708137512207, "genuineness": 4.042920112609863, "identity": 0.5081769227981567, "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/emo_Pain/94bb1c212cf7088b3ad33acd65c085ddce65db00a8508347d8aa3bb1dbac26b5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "94d10ade68224b5c88b6d044cb2c069ff30cb98dbe9cc265e177b229faca6b42", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7d1509666d05fe6dc2216b21fbe9b6cf0602f17138ec022e784d3808113cd2fc", "generation_seconds_batch": 23.225750015000813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3468351364135742, "target_z": 0.15624996761583532, "ce": 5.569881439208984, "genuineness": 4.469690799713135, "identity": 0.57351154088974, "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/emo_Pain/94d10ade68224b5c88b6d044cb2c069ff30cb98dbe9cc265e177b229faca6b42.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "94eb5ff647ed9e20e66b56acbcefd1187f9e2451adf468f3f0dc62ba2ba556f6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "87cb9a21c4f77f02e9f54e166ccf3094e008574e7bdde0bb88f15be788eeadff", "generation_seconds_batch": 21.80773731006775, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.005327016115188599, "target_z": -0.43750001295366586, "ce": 6.125868320465088, "genuineness": 4.268857479095459, "identity": 0.5019178986549377, "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/emo_Pain/94eb5ff647ed9e20e66b56acbcefd1187f9e2451adf468f3f0dc62ba2ba556f6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "94ebdffe025f99782f9736f33bb9652be543bde6cf424fa804f2853a21bb772e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9734a2eadbd3c4a022f8a6e37fd311a54a5323989e6a6a006e03ce85736f5e39", "generation_seconds_batch": 23.28106885298621, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.348263263702393, "genuineness": 4.632867813110352, "identity": 0.4402611255645752, "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/emo_Pain/94ebdffe025f99782f9736f33bb9652be543bde6cf424fa804f2853a21bb772e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "95620c1a9197cc17be71ec0fd50a736c0a6b55af41f0018ad45126e158f240b3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "52eadbe708756e7a9c3f46a3229f1056027e56187f88c572ca8154bf2998d8bd", "generation_seconds_batch": 22.23442521202378, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.133810997009277, "genuineness": 4.20609712600708, "identity": 0.5849039554595947, "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/emo_Pain/95620c1a9197cc17be71ec0fd50a736c0a6b55af41f0018ad45126e158f240b3.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": "95691b71aed1bb43c34a31576e5928f62ad79e65db423c05b3eeca94b8f33a7f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "78b6f6f8409535f08f461f7258313c13a203e90dc121ddf0e217441eb92820c8", "generation_seconds_batch": 23.474090890958905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44344601035118103, "target_z": 0.3242187540480206, "ce": 5.788304805755615, "genuineness": 4.337893962860107, "identity": 0.7651892900466919, "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/emo_Pain/95691b71aed1bb43c34a31576e5928f62ad79e65db423c05b3eeca94b8f33a7f.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": "959fc13ce3e3633d81ba3bb8e9e9742f6b41715c2ec54f2d5109f433d935380f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bcf852939c685108b66e8765d6c23ccbf9ae7f195703f83ab63afa613b1e967e", "generation_seconds_batch": 22.05796927504707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5960359573364258, "target_z": 2.3281250097152495, "ce": 5.593709468841553, "genuineness": 4.457139015197754, "identity": 0.7171674370765686, "hypothesis": "Deine Nachricht kam, als sie schon weg war. Ich lass sie zweimal. Dann legte ich das Handy weg. Sag 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/emo_Pain/959fc13ce3e3633d81ba3bb8e9e9742f6b41715c2ec54f2d5109f433d935380f.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": "95e8ebc98b6e4160ad76de9d69ceb58c6626d6a45cf636a4085fd8f4dfb5f53a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "57ebcea26a2995a0ea9c480c24756bfc008d9884f3f6c2d6ae40eee4b60bce64", "generation_seconds_batch": 21.07561269099824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9399808645248413, "target_z": 1.187500012953666, "ce": 5.863759994506836, "genuineness": 4.137060642242432, "identity": 0.6447544097900391, "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/emo_Pain/95e8ebc98b6e4160ad76de9d69ceb58c6626d6a45cf636a4085fd8f4dfb5f53a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "964d7cc0b8f0f41a9122faf58111669018e99616b34eb332fe04305897248796", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "af7fd91e96437d64eaa2239eb3a868144b03b2d51acc39d1413aee5b592dc2fd", "generation_seconds_batch": 23.917818334070034, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20051464438438416, "target_z": -0.09814453762563242, "ce": 6.157639026641846, "genuineness": 4.281409740447998, "identity": 0.573945164680481, "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/emo_Pain/964d7cc0b8f0f41a9122faf58111669018e99616b34eb332fe04305897248796.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "96984a34db33c7ee39170fd55ebd9e986d1165308667a301980d47c7684a5347", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4600537f41ef1880d53d63cf1bda7519e1d0a1d4112e815e50fe7e53fc95bf2b", "generation_seconds_batch": 17.747265412006527, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.919569969177246, "target_z": 2.8906249967615834, "ce": 5.80021858215332, "genuineness": 4.21237325668335, "identity": -0.09178082644939423, "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/emo_Pain/96984a34db33c7ee39170fd55ebd9e986d1165308667a301980d47c7684a5347.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "970071a8139a94dd39cade136344f72077953cbded65a52852b947658dfad7db", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9dbd26596da6b755ab3c1e534c11b3bc301ba5b1f740cc153f243837321b121a", "generation_seconds_batch": 23.908837343915366, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2575262784957886, "target_z": 0.0009765509631413283, "ce": 6.2767791748046875, "genuineness": 4.507347106933594, "identity": 0.6332640647888184, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/970071a8139a94dd39cade136344f72077953cbded65a52852b947658dfad7db.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "97cf53b8719ddfb65f7b434b3954a4690d0539d5976bfe767045c40b40e88b77", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0a0184326fc543ca8f4d496c22296a362e8fe5c6c904be41d09e0f86c3ce2e89", "generation_seconds_batch": 24.106327605899423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0907040536403656, "target_z": -0.2890625048576247, "ce": 6.070269584655762, "genuineness": 4.413206577301025, "identity": 0.6342109441757202, "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/emo_Pain/97cf53b8719ddfb65f7b434b3954a4690d0539d5976bfe767045c40b40e88b77.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": "97e479601ae5cb16f8b0fc76ad2de96e6f9f8e894535876649c261ba51702429", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Neem ze mee als je weggaat.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De sleutels liggen op de keukentafel.\n(Natural clear speech.) [3.8 seconds duration] Ik keek in de la... en daarna naast het raam.\n(Natural clear speech.) [3.2 seconds duration] Neem ze mee als je weggaat.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "25e716b73eadd990f46d192619e58fa83e358ef831e064e42bc2246c37526372", "generation_seconds_batch": 18.02509222610388, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2311268150806427, "target_z": -0.044921862451136184, "ce": 6.062326908111572, "genuineness": 4.369274139404297, "identity": 0.22299331426620483, "hypothesis": "De sleutels liggen op de kuikentafel. Keek in de laag en daarna naast het raam, neem ze mee als je weggaat.", "wer": 0.13636363636363635, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/97e479601ae5cb16f8b0fc76ad2de96e6f9f8e894535876649c261ba51702429.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": "97e4d89836c6af44530fad0aa380d6d4fedaefa31c4ca92fd644658b301e00a5", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4582190a01fb68791b74d79ffea4f811e2d61a30d665565aca74fd42c293c101", "generation_seconds_batch": 16.699873942066915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5962259769439697, "target_z": 0.5898437767169359, "ce": 5.530168056488037, "genuineness": 4.557555198669434, "identity": 0.7581748962402344, "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/emo_Pain/97e4d89836c6af44530fad0aa380d6d4fedaefa31c4ca92fd644658b301e00a5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "983b40181f07b5b7f01129b2b8a54b5142720d7d2ac636c2ac1e17773475ba83", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "accd102f9aa587620fdf4f219f83de7f8d3615055f96de571f7379706bdb8d5d", "generation_seconds_batch": 21.652912486926652, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.2767791748046875, "genuineness": 4.13078498840332, "identity": 0.3275096118450165, "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/emo_Pain/983b40181f07b5b7f01129b2b8a54b5142720d7d2ac636c2ac1e17773475ba83.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "98e1676d37e7f4b562e3e5c31082650c651ca22b4d26d508867d112c059a2b1b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7498311dcc9e61fddc1a6a79f3d484b634e6ff9056fe135e2b0744a0640089d8", "generation_seconds_batch": 22.226331147016026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11373338103294373, "target_z": -0.24902344903685866, "ce": 6.205295085906982, "genuineness": 4.2751336097717285, "identity": 0.3747548460960388, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/98e1676d37e7f4b562e3e5c31082650c651ca22b4d26d508867d112c059a2b1b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "9924bd550ac17b139a1094f4535dffc7e2d498f30ff7c0a89bd3919d74dcc584", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6616f5fdb32f65090832adbea9d8d728202c8c884b1ea5a015b8fc511e8db49a", "generation_seconds_batch": 18.172874436015263, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2732536196708679, "target_z": 0.028320288819079567, "ce": 6.157639026641846, "genuineness": 4.482243061065674, "identity": 0.45958462357521057, "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/emo_Pain/9924bd550ac17b139a1094f4535dffc7e2d498f30ff7c0a89bd3919d74dcc584.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "7126e615db4b27aba6ceb0ec23f1e22159a8639bb39b70a3f23a98c0a3a6cc7c", "generation_seconds_batch": 17.93150353396777, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.308549880981445, "genuineness": 4.6203155517578125, "identity": 0.564405083656311, "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/emo_Pain/99bf1b7d3eed7b885fdae888ab00b4960791c3971d259b46f771306d2801e39b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "99f0eedbb55f0fcde0db5cdf4d2c32da2600cb0f6bdf0b88f7c1143eab39ae4f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "1c10f12643e49d237248bc360778929128331b5dc5ec1c2b528e3a4e0cab430b", "generation_seconds_batch": 23.51374213292729, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.232060194015503, "target_z": 1.6953125242881235, "ce": 5.839932441711426, "genuineness": 4.557555198669434, "identity": 0.6474512815475464, "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/emo_Pain/99f0eedbb55f0fcde0db5cdf4d2c32da2600cb0f6bdf0b88f7c1143eab39ae4f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "9a14530aff5461099e614bc4222cc2150c0ce1bf3ce489a32932a00c9cf3e7d9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "bc51806caea781828249cb17fe86c11b3e78ea8297fc90b5d63264b4e6e204a1", "generation_seconds_batch": 16.084019648027606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7512526512145996, "target_z": 0.859374951423753, "ce": 6.117925643920898, "genuineness": 4.595211505889893, "identity": 0.7997174859046936, "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/emo_Pain/9a14530aff5461099e614bc4222cc2150c0ce1bf3ce489a32932a00c9cf3e7d9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "9a4d1d98fa378900db6b4dd93a18d86c29745885063617f65c02bf6c513dd3a4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "5ec74fe094777286a58fb9fd867d414ce4cbe0099c066277ee04e78cfeb10925", "generation_seconds_batch": 22.429160495987162, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0008334815502166748, "target_z": -0.4453125242881235, "ce": 5.967014789581299, "genuineness": 4.319066047668457, "identity": 0.710229754447937, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer 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/emo_Pain/9a4d1d98fa378900db6b4dd93a18d86c29745885063617f65c02bf6c513dd3a4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "9b5b735e058c582b2c39c4d14540a7cd7bbe47838bee2b1f06a8913333f89a5f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4128e4dd29337a2fc05ec8699f901adc03c60de5090e058572a334eb30f1d520", "generation_seconds_batch": 22.548240453004837, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4524330496788025, "target_z": 0.3398437249022724, "ce": 5.77639102935791, "genuineness": 4.7081804275512695, "identity": 0.5840302109718323, "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/emo_Pain/9b5b735e058c582b2c39c4d14540a7cd7bbe47838bee2b1f06a8913333f89a5f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "9b93acb8c03835a5fe13f33912cc76681681842c25e4d926746632c1f12bb11d", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f5ee8cf0b82adde0bae280d09bd7f762cc8f3880418d63db4532f57bce4484c1", "generation_seconds_batch": 17.438399091013707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11092492938041687, "target_z": -0.2539062556672288, "ce": 5.855817794799805, "genuineness": 4.582659721374512, "identity": 0.7387924790382385, "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/emo_Pain/9b93acb8c03835a5fe13f33912cc76681681842c25e4d926746632c1f12bb11d.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "62e199570d6cccfb0584d4dec0cb5be36ffd1642104d3ebde148268c11d1e863", "generation_seconds_batch": 16.847944028908387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.284721851348877, "genuineness": 4.375550270080566, "identity": 0.6836303472518921, "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/emo_Pain/9c09e940c5dab53287ffae2d1a8fa5296f41b2df71714656aaa2bb8830bf5162.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": "9c0b4d379ca3b707c39833ad9ccbba2515707681609f82c95e590d2d6e609290", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0431c94aba944115b36d1b5f3daac2eebf1ce6f77a980c59ba1b7f7a3f57f6a6", "generation_seconds_batch": 17.29066447902005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3962639570236206, "target_z": 0.24218748866554235, "ce": 5.903473377227783, "genuineness": 4.595211505889893, "identity": 0.8972874283790588, "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/emo_Pain/9c0b4d379ca3b707c39833ad9ccbba2515707681609f82c95e590d2d6e609290.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "9c17ff883e63260825a83aac4f423cf58dc634b54cee70a9ef7f29f68c66459b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "fb25d0a00fa71ca74a75f2038b2ffacb2f958b988fa002f573afbdd241e5f5ae", "generation_seconds_batch": 23.50753080798313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5894856452941895, "target_z": 0.5781249579005859, "ce": 5.736677646636963, "genuineness": 4.557555198669434, "identity": 0.5593645572662354, "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/emo_Pain/9c17ff883e63260825a83aac4f423cf58dc634b54cee70a9ef7f29f68c66459b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "9c2595279608e0a4a2607846d50e5153eed903d3d1f55233660922d63059b41e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "a6728ad2c29bddc19f269f1f814305a063d06a68df81274ddee0336ef15457f5", "generation_seconds_batch": 22.834977486985736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5600876808166504, "target_z": 2.2656249190395883, "ce": 6.133810997009277, "genuineness": 4.306513786315918, "identity": 0.5865153074264526, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/9c2595279608e0a4a2607846d50e5153eed903d3d1f55233660922d63059b41e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "9c25d6db1afcfb12b7535c5336a80bdf8b5f69afc7acac5e5c283fce76819e33", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f64814d71a35f81da1fa16de5411bd71607ed0acb5a3688e7153aa6ec75d2c4f", "generation_seconds_batch": 23.45004660298582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04464539885520935, "target_z": -0.36914061649915675, "ce": 6.459460258483887, "genuineness": 4.6956281661987305, "identity": 0.6820147037506104, "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/emo_Pain/9c25d6db1afcfb12b7535c5336a80bdf8b5f69afc7acac5e5c283fce76819e33.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "9c532a3938571d559bf040cce29af0666b1bc3e19270a595f55b1388a950386c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "22c5524481ad54560a2ecc2fcb4f0d68d076e6824a3e241f814fd9f52e9d948e", "generation_seconds_batch": 22.86629430402536, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4792041778564453, "target_z": 2.124999922278005, "ce": 5.6334228515625, "genuineness": 4.093128204345703, "identity": 0.7162783145904541, "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/emo_Pain/9c532a3938571d559bf040cce29af0666b1bc3e19270a595f55b1388a950386c.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "e288ec3abbe4bd183fea1dea200372a127140166038ab79cc98fc5a33756280a", "generation_seconds_batch": 16.65194062003866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.055879220366477966, "target_z": -0.34960936407034443, "ce": 6.062326908111572, "genuineness": 4.068024158477783, "identity": 0.6394271850585938, "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/emo_Pain/9c969db20a6db380e08592f6aed9c9189b3db634d677ebe8f573f593ebdfd2d3.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": "9ca8253327faf752bd7138e60da4c5e45766306f3874aca89ac8f09bf37fe132", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "6bfc3a4f66a6d288dec6852a9641468bd2d185b00fba9800eecd26458cd76ca7", "generation_seconds_batch": 17.369245629990473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 6.09409761428833, "genuineness": 4.7458367347717285, "identity": 0.8166667222976685, "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/emo_Pain/9ca8253327faf752bd7138e60da4c5e45766306f3874aca89ac8f09bf37fe132.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "9cc3381c5968ef6bd3318b9f722646f711df1525b370e7ea2781599401c48c8e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "69858caa0535e199b7fcd1749e7f4e108cbdb1c1ef0b96b7eaf01bb58bf5099d", "generation_seconds_batch": 22.43720601592213, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6097065210342407, "target_z": 0.6132812070909818, "ce": 5.95112943649292, "genuineness": 4.118232727050781, "identity": 0.4040365517139435, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/9cc3381c5968ef6bd3318b9f722646f711df1525b370e7ea2781599401c48c8e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "9d0b13e00d793a39174609743831efd7c0b64a8ceca72c62afef779438da2248", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9cfd21c52e65074d5ea326f994c19f1a624418ad763a7fa4d6db2d7cacd1f5b4", "generation_seconds_batch": 22.30763688602019, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19967210292816162, "target_z": -0.09960938997767617, "ce": 6.205295085906982, "genuineness": 4.545003414154053, "identity": 0.46696168184280396, "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/emo_Pain/9d0b13e00d793a39174609743831efd7c0b64a8ceca72c62afef779438da2248.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "9d138babacf86f8305818fdd5f0c260206ea6b53c785a3f19aa219d718750d86", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5763017504e079d5376d4199de4a7b6071e9abc9e5484eeec59f3582be372003", "generation_seconds_batch": 21.739319278975017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.039028480648994446, "target_z": -0.3789062556672288, "ce": 5.871703147888184, "genuineness": 4.093128204345703, "identity": 0.6431064605712891, "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/emo_Pain/9d138babacf86f8305818fdd5f0c260206ea6b53c785a3f19aa219d718750d86.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": "9d6bfde8ad8f37408362b59611c9a9997916a5e247cba737f33b16b8d532efed", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a6cdbce22eb9af94e8546304be276fbbaccac2d079e08ffa924aac5bea1f32ba", "generation_seconds_batch": 21.540408671018668, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.2251298427581787, "target_z": 3.4218749384700873, "ce": 5.327630043029785, "genuineness": 4.18099308013916, "identity": 0.5601577758789062, "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/emo_Pain/9d6bfde8ad8f37408362b59611c9a9997916a5e247cba737f33b16b8d532efed.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": "9dfad38af335e4155508dcf631994b011b5ccaaa23ab31b03a05d879a3ac6716", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c7918cdae4dd833ed9df6488a2bab2bdc1d739a1fbe3826ff46b8a98896d9467", "generation_seconds_batch": 22.614626239053905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.587048888206482, "target_z": 2.3124999870463343, "ce": 5.649308204650879, "genuineness": 4.243753433227539, "identity": 0.5853208303451538, "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/emo_Pain/9dfad38af335e4155508dcf631994b011b5ccaaa23ab31b03a05d879a3ac6716.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": "9e99de41474818e0ee976f5c204df6bfd8abbe38d1f7c18772b79895bf57fa43", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "f5082584bd31aed4ab274f21e08d682db917c5e6f60b2927f1697bf5b095c605", "generation_seconds_batch": 25.231766785960644, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3488918542861938, "target_z": 1.898437404466714, "ce": 5.8875885009765625, "genuineness": 4.394378662109375, "identity": 0.6398921012878418, "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": 0.03999999999999915, "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/emo_Pain/9e99de41474818e0ee976f5c204df6bfd8abbe38d1f7c18772b79895bf57fa43.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "9f4f01783b035f03ed0be95c443413e1dd04a1bc9f7164ec34a464a737a21f8c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b830e64691a9bcadc736c36d3794174068c0c09bbaa40130673ae5109862dc62", "generation_seconds_batch": 22.397373902029358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09744435548782349, "target_z": -0.27734373785593824, "ce": 5.911416053771973, "genuineness": 4.871357440948486, "identity": 0.6802711486816406, "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/emo_Pain/9f4f01783b035f03ed0be95c443413e1dd04a1bc9f7164ec34a464a737a21f8c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "9fdffe429095fffdd4eabba16bea2e8f572cea0a46d3f986eebe9485c0ecb8a8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f200fb04bff05e609b72552db2bb67716153aca62dc4e54eaab2046b6f0d3b4c", "generation_seconds_batch": 24.38280466804281, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059249356389045715, "target_z": -0.34375000647683296, "ce": 6.125868320465088, "genuineness": 4.3504462242126465, "identity": 0.4670614004135132, "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/emo_Pain/9fdffe429095fffdd4eabba16bea2e8f572cea0a46d3f986eebe9485c0ecb8a8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "9fe8fbda07e059c705774231a4f0533df52fc8cca7f8a147c3c6a5283dd5f994", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "829f7847c33730fe73690fa1563a56562d46e41169b1428fd163526fd8c66c5c", "generation_seconds_batch": 17.27680012001656, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36761772632598877, "target_z": 0.19238281958403602, "ce": 5.8875885009765625, "genuineness": 4.168440818786621, "identity": 0.3793087601661682, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/9fe8fbda07e059c705774231a4f0533df52fc8cca7f8a147c3c6a5283dd5f994.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "a03d21f348cf99320ebc7514bc5cca5bd13a8144b17cfcb66f6133a5b9a09b73", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8a9fdd0dfbca1e633c43e26d3274ee291f7fe114035cfe03107684cb25598fe0", "generation_seconds_batch": 23.526632967987098, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.324435234069824, "genuineness": 4.632867813110352, "identity": 0.6354824304580688, "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/emo_Pain/a03d21f348cf99320ebc7514bc5cca5bd13a8144b17cfcb66f6133a5b9a09b73.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "a05c3e7a7bd1de29ac99df6c92305bd77095bcf8281f2ba827da8502d0df0d9a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5f34a4b2d269bf8f1ed1649d72a7e3a2a55a4b281c63bb3266df6bb4ddce831c", "generation_seconds_batch": 27.929668414988555, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8590973615646362, "target_z": 1.0468750161920823, "ce": 6.157639026641846, "genuineness": 4.7834930419921875, "identity": 0.7386384606361389, "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/emo_Pain/a05c3e7a7bd1de29ac99df6c92305bd77095bcf8281f2ba827da8502d0df0d9a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "a06e4839358b91d2b03ef323fba2bca221d9cc2760afb3fc1808ead561130e83", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8ae2c22ec444d37cc1b09c26297b77e63f12370278014e48eb963ba45f53ef8e", "generation_seconds_batch": 21.73654729896225, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26904094219207764, "target_z": 0.020996078873524344, "ce": 6.117925643920898, "genuineness": 4.846253395080566, "identity": 0.7053074240684509, "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/emo_Pain/a06e4839358b91d2b03ef323fba2bca221d9cc2760afb3fc1808ead561130e83.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "a07de0c38729d69ec4091fc6f7d96089baccae62bf5c60a8ef4e62398b5df9a7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0b1c577b3610459d5bc7acc096883180b2c6e935a6ae9b042110ed54fb2fe91c", "generation_seconds_batch": 23.447647182969376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.07478666305542, "target_z": 1.421874938470087, "ce": 5.8875885009765625, "genuineness": 4.369274139404297, "identity": 0.5175583362579346, "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/emo_Pain/a07de0c38729d69ec4091fc6f7d96089baccae62bf5c60a8ef4e62398b5df9a7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "a087657c578d41cb7c61c4bb838ad0bb1c7875263e63715825309580701538c4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8df33f25c3b690c3a4a380ac2c5b09810bacd6ab977324c68df2f26e270c2daf", "generation_seconds_batch": 21.966157326009125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 5.919358730316162, "genuineness": 4.532451152801514, "identity": 0.707774817943573, "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/emo_Pain/a087657c578d41cb7c61c4bb838ad0bb1c7875263e63715825309580701538c4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "a0e99adb0211182d3935fd57d4b53d7312f7c987b753fcddfbb723843bb6c812", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "10f6a090bcdcf4a735acac1843d19032e46680e60596a6b2c80c00e7bdcb0e05", "generation_seconds_batch": 16.501453459030017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11766523122787476, "target_z": -0.24218748866554235, "ce": 5.935244083404541, "genuineness": 4.959221839904785, "identity": 0.6457144021987915, "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/emo_Pain/a0e99adb0211182d3935fd57d4b53d7312f7c987b753fcddfbb723843bb6c812.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "a12a92283f26fc12f7afb0fd65ff952c838f783d7f896fa816a34d96348d8149", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "078bb06bb2800214ecea0a3f0be372c2a509758fa8c2ed356a5ed5a9af95e3c4", "generation_seconds_batch": 21.723850366077386, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.29515957832336426, "target_z": 0.06640624271356295, "ce": 5.692992687225342, "genuineness": 4.482243061065674, "identity": 0.8007977604866028, "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/emo_Pain/a12a92283f26fc12f7afb0fd65ff952c838f783d7f896fa816a34d96348d8149.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "a12b3a627ec3c2b40d9b813cd309e0e63b35c52a53be75470d10fbce40019353", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e32249d5f162da3afc7af9298a90bbba2edcc6f8de70304c22470442069c97e4", "generation_seconds_batch": 24.106327605899423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11092492938041687, "target_z": -0.2539062556672288, "ce": 6.133810997009277, "genuineness": 4.532451152801514, "identity": 0.3830786943435669, "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/emo_Pain/a12b3a627ec3c2b40d9b813cd309e0e63b35c52a53be75470d10fbce40019353.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "a154a148a5979c2c9bd1cb098be89beec0b7ef2634adc40b0cef3714e3a4a6dc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "edc7e91010728a916f7159bd859449387c85a8cdc7659c85a1072cb7785b31d7", "generation_seconds_batch": 24.084413825999945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0963209718465805, "target_z": -0.27929686568955264, "ce": 6.24500846862793, "genuineness": 4.18099308013916, "identity": 0.4733947515487671, "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/emo_Pain/a154a148a5979c2c9bd1cb098be89beec0b7ef2634adc40b0cef3714e3a4a6dc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "a17a9880768d43c2d5643157628d6a403eb68183dc29fb5d3b77a64ca3132ab8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "235a9d509572423dfd06880cd6244635b6196827e7c7e6575cb9faea663b4d4b", "generation_seconds_batch": 16.31760609499179, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.36256250739097595, "target_z": 0.18359375728643706, "ce": 6.268836498260498, "genuineness": 4.557555198669434, "identity": 0.4162479639053345, "hypothesis": "The door's still open. I put your coat on the chair. Tell me when you are ready", "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/emo_Pain/a17a9880768d43c2d5643157628d6a403eb68183dc29fb5d3b77a64ca3132ab8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "a1ac9582f458d354ac409cbe5924c6e845437eb7c7d8f9abb194f85b9af64518", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "960a89944a339cd81522877339b5cf067a6ca68657d0d743ec147dca1d200d97", "generation_seconds_batch": 21.79103117901832, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17271094024181366, "target_z": -0.14648438026242677, "ce": 5.919358730316162, "genuineness": 4.821148872375488, "identity": 0.6148220896720886, "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/emo_Pain/a1ac9582f458d354ac409cbe5924c6e845437eb7c7d8f9abb194f85b9af64518.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": "a29d36766d659438bde74b0f16e7b07c3ea7b34e8848c09ebddbdf5cec2f244a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8d1500fae9e030332871ba9ea9de112a715ced7a4b7636d75a8b6bfbf5d377ee", "generation_seconds_batch": 26.607766156084836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 5.689021587371826, "genuineness": 3.998987913131714, "identity": 0.5747512578964233, "hypothesis": "The keys are on the kitchen table. I I checked the drawer, then looked beside the window. Take them with you when you leave.", "wer": 0.043478260869565216, "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/emo_Pain/a29d36766d659438bde74b0f16e7b07c3ea7b34e8848c09ebddbdf5cec2f244a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "a2ba0be85a148810d213424ee6d0dd7a1fb96206063cee00ab6fcb602aee730f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8906a9ab051cdc1741020523b187dff2a68237848b91d53b191af89c6bb0e1f1", "generation_seconds_batch": 23.23650970601011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07385332882404327, "target_z": -0.31835937054717733, "ce": 6.252951145172119, "genuineness": 4.507347106933594, "identity": 0.8367712497711182, "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/emo_Pain/a2ba0be85a148810d213424ee6d0dd7a1fb96206063cee00ab6fcb602aee730f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "a30362fa56f9fec34ad0c99b54f64059642bdb81fef3f1c4958d6904c9e041da", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "901feadd984b41b1c085ecb2d468a4a14d5fd5fef3bc7173a06f9d7e6b3fe73c", "generation_seconds_batch": 23.821633242070675, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.065799593925476, "target_z": 1.4062499158011719, "ce": 5.661221981048584, "genuineness": 4.024092197418213, "identity": 0.4634495675563812, "hypothesis": "Ik heb het 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/emo_Pain/a30362fa56f9fec34ad0c99b54f64059642bdb81fef3f1c4958d6904c9e041da.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "a334f2e59d17c3f16b6ade4622152786a63c535999546c544ad800bad9a1f54c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "48e4c83875274799f0e53cf8aaceb23b49cdf9c36dae9fa4c974ec9dc8518ffd", "generation_seconds_batch": 26.986887359060347, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10081449151039124, "target_z": -0.27148438026242677, "ce": 5.839932441711426, "genuineness": 4.6705241203308105, "identity": 0.7375268340110779, "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/emo_Pain/a334f2e59d17c3f16b6ade4622152786a63c535999546c544ad800bad9a1f54c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "a3d5f06c7f6adffc17d2adab95bbcbcd75b2864b5f18e14683b7e6ddbea2515b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3ff3e804a5d6a7772ec2aed7937f49970884ce9121df178c6b8f24b7cd0dca1f", "generation_seconds_batch": 16.560726227005944, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.157639026641846, "genuineness": 4.796044826507568, "identity": 0.7328808903694153, "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/emo_Pain/a3d5f06c7f6adffc17d2adab95bbcbcd75b2864b5f18e14683b7e6ddbea2515b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "a4633b55a975790c0c3c99325109abcc61591f25c5cf3f30095a98c957ad7a5c", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "efc69ff51f0ad091864113f693e4ca3bbe52e81c159320ce34d3f3e6134b55e6", "generation_seconds_batch": 18.678087811917067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0042036473751068115, "target_z": -0.43945311487994854, "ce": 6.038498878479004, "genuineness": 4.262581825256348, "identity": 0.6912922263145447, "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/emo_Pain/a4633b55a975790c0c3c99325109abcc61591f25c5cf3f30095a98c957ad7a5c.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": "a501828f8402196740054728561b1034f2fd7e1b79ab9d091dc5f721046ae46b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0c8b8749f137645eb84571b7ef322508b80cd1b1ebf6dba753bab4078fbd4c9d", "generation_seconds_batch": 22.488499677041546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0837737321853638, "target_z": 1.4374999611390025, "ce": 5.6532793045043945, "genuineness": 4.068024158477783, "identity": 0.34262531995773315, "hypothesis": "De wegjat gaat linksaf. Naar de brug. Loop 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/emo_Pain/a501828f8402196740054728561b1034f2fd7e1b79ab9d091dc5f721046ae46b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "a5cf07a9d3fa38d928c53f3f90c516c7eb99796928060bea5921892135ee8b36", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "24fce40347a8c8111f3dcc88a6ed1ed121cf3bfbad745005f45b7f945acbf61f", "generation_seconds_batch": 23.04837839002721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.48501116037368774, "target_z": 0.3964844061697585, "ce": 5.935244083404541, "genuineness": 4.582659721374512, "identity": 0.47664710879325867, "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/emo_Pain/a5cf07a9d3fa38d928c53f3f90c516c7eb99796928060bea5921892135ee8b36.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "a64ea755d3e47b65a5616ba9a191632a48ea2bc8dc4ad508547fa34f2b6433da", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8dae89e9dd3599c5e108670a0d35c148590d2ba16006ca121638d50d3987c6d6", "generation_seconds_batch": 23.555264313006774, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2904759645462036, "target_z": 1.7968748607480918, "ce": 5.7247633934021, "genuineness": 4.432034492492676, "identity": 0.4134296178817749, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het augen 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/emo_Pain/a64ea755d3e47b65a5616ba9a191632a48ea2bc8dc4ad508547fa34f2b6433da.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "a685f8578b735b67d7e4336a2fdc9a13cf9e64eb1453e12d6021aa2be71bdcdb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8d812e4e56de1a0cbbac2c3eab53b3fc822ee90ef452634f0a1a2b474f796307", "generation_seconds_batch": 18.47566448105499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07497671246528625, "target_z": -0.31640624271356294, "ce": 6.173524379730225, "genuineness": 4.557555198669434, "identity": 0.562529444694519, "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/emo_Pain/a685f8578b735b67d7e4336a2fdc9a13cf9e64eb1453e12d6021aa2be71bdcdb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "a71b5e7b94362dd960da0ce4f71f15593e72d0e767ee419c126d3158af6891c3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ccc8c6e46580d8463618cbc3f2da02764efeb4c0b38d678b60360544ef0afb9b", "generation_seconds_batch": 18.37425973196514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4513096809387207, "target_z": 0.3378906229759897, "ce": 5.824047088623047, "genuineness": 4.7332844734191895, "identity": 0.6716189980506897, "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/emo_Pain/a71b5e7b94362dd960da0ce4f71f15593e72d0e767ee419c126d3158af6891c3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "a79dc90f7dbf464d6155b4200776572f762bffef5645d3182f167de06124f05f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5049fd68d96cc9deac3741b002f667b0497da37cf768d42eee66cd65d695d419", "generation_seconds_batch": 22.476641788030975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5151524543762207, "target_z": 2.1875000129536657, "ce": 5.200547218322754, "genuineness": 4.243753433227539, "identity": 0.3549841344356537, "hypothesis": "La porte est encore ouverte, j'ai posé ton tour 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/manual-comparison-v1/audio/emo_Pain/a79dc90f7dbf464d6155b4200776572f762bffef5645d3182f167de06124f05f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "a7a94177b70ad0adab9abb9f19201e8b53aa4d5f99a79deafea12c26ec30cba3", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "caaa875590e0c02df2ccf0c8daec02bf186524c5f6c7dc14db642db4c7845cba", "generation_seconds_batch": 16.573090917081572, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.26173895597457886, "target_z": 0.008300760908696553, "ce": 5.95112943649292, "genuineness": 4.532451152801514, "identity": 0.5851609110832214, "hypothesis": "Ich hab die kleine Schachtel gefunden. Sie stand neben dem alten Fenster. Wir können gemeinsam hineinschauen.", "wer": 0.0625, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/a7a94177b70ad0adab9abb9f19201e8b53aa4d5f99a79deafea12c26ec30cba3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "a8384b239465b25709ed5453cbbfbac2a7f2720010f3b9dffc699d6ecd410178", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5abaeb3f5d7f0232084b663ad003311f2d499eaacf8a634fba5e6efb442669c6", "generation_seconds_batch": 23.39202202996239, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5288230180740356, "target_z": 0.47265621032939825, "ce": 6.09409761428833, "genuineness": 4.595211505889893, "identity": 0.40262269973754883, "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/emo_Pain/a8384b239465b25709ed5453cbbfbac2a7f2720010f3b9dffc699d6ecd410178.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "a84f6b0674b189200706cace6520c1de6dda61789b9e8f6a6dd8b37651a75b0b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7837d0c78b318acfd013fb746daef9d7ee494c0d625ddb3a730f4632663e6d86", "generation_seconds_batch": 18.860914860037155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44119924306869507, "target_z": 0.3203124983807918, "ce": 5.649308204650879, "genuineness": 4.582659721374512, "identity": 0.5603808760643005, "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/emo_Pain/a84f6b0674b189200706cace6520c1de6dda61789b9e8f6a6dd8b37651a75b0b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "a854432b280d5fdabb61c33a9cf51741cca4602fd41ad8bfe6d05dfbef36d303", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1bc0511099d5c86d4d803d16d16ce765a233b6142db3c62bc1c5996be00f3da0", "generation_seconds_batch": 23.16149537707679, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0882673263549805, "target_z": 1.445312576102787, "ce": 5.214447021484375, "genuineness": 4.218649387359619, "identity": 0.3173607587814331, "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/emo_Pain/a854432b280d5fdabb61c33a9cf51741cca4602fd41ad8bfe6d05dfbef36d303.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "1fd8cb455e883483ec83f8a7e95c9796fdc619a971d220c610f25cd8d3c980a1", "generation_seconds_batch": 16.749554385081865, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0042036473751068115, "target_z": -0.43945311487994854, "ce": 6.117925643920898, "genuineness": 4.243753433227539, "identity": 0.694062352180481, "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/emo_Pain/a8bd001b1fcafd2f49b65d73e6ed12567e4c9664909a4d08700c9febfd977540.mp3" }, { "id": "heldout-fr-0", "language": "fr", "sentences": [ "Ton message est arrivé après mon départ.", "Je l’ai relu... puis j’ai rangé mon téléphone.", "Dis-moi quand nous pourrons parler." ], "text": "Ton message est arrivé après mon départ. Je l’ai relu... puis j’ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-fr-0", "key": "a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea", "bucket": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "49e9a0a880cb2c7068ad44b2ed37255a09c0354c6d4019b11c5108b05e7a7743", "generation_seconds_batch": 18.21413619199302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07048317790031433, "target_z": -0.3242187540480206, "ce": 6.197352409362793, "genuineness": 4.482243061065674, "identity": 0.7773113250732422, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi, quand nous pourrons parler ?", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/a94b7d5da78ff217f8e5a4f7ae4520900bef0ad764d0e4ee898e69de10c28aea.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": "a980bf68a9c0ec20505de7e873f3e5b603be4a01202e49977e8ac959dedd91f9", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bcb3e43d7f7321bfa9316d6a71708fed92969b7b4b2398f861c174a6087638a6", "generation_seconds_batch": 17.037321664043702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1575452834367752, "target_z": -0.17285156715522368, "ce": 5.935244083404541, "genuineness": 4.413206577301025, "identity": 0.7965949177742004, "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/emo_Pain/a980bf68a9c0ec20505de7e873f3e5b603be4a01202e49977e8ac959dedd91f9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "a9b7e81f757c45770fb8ed54d359544bcc137a4d699e51e43651fba77b2dc114", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "90fe35ab4f988f589ada0cc05d1f9d81bb5c0c5f8a9785c867eef14832597f17", "generation_seconds_batch": 22.74764752702322, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08845728635787964, "target_z": -0.29296876052485354, "ce": 6.205295085906982, "genuineness": 4.432034492492676, "identity": 0.41685807704925537, "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/emo_Pain/a9b7e81f757c45770fb8ed54d359544bcc137a4d699e51e43651fba77b2dc114.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "a9d5d133206f736b88717fff9a65b4464cd97079835812fa6356afef043adb6a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d51fd7868fdae5a953c78bf9318ee59c2b1cc32e63da8f80c57ebaece6613cb3", "generation_seconds_batch": 23.031720163999125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.4228451251983643, "target_z": 3.765625022668915, "ce": 5.839932441711426, "genuineness": 4.162164688110352, "identity": 0.11311649531126022, "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/emo_Pain/a9d5d133206f736b88717fff9a65b4464cd97079835812fa6356afef043adb6a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "aa833429dd4ab5612d43ac910f04f30f77a1d3a3507907a6ef15ba5f2ccd5672", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7b7bfde15e47eeabf9010b0af693e4169e0d2e1b5c0ca7a005a1c6afbcdd14f2", "generation_seconds_batch": 27.175588411046192, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9040325880050659, "target_z": 1.1249999222780047, "ce": 6.125868320465088, "genuineness": 4.457139015197754, "identity": 0.5183210372924805, "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/emo_Pain/aa833429dd4ab5612d43ac910f04f30f77a1d3a3507907a6ef15ba5f2ccd5672.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "aa971def5ae7062694d490a1f8f4d96e49b3fa0451ab580fed910361f41020f2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3253effc0fe72834a4db1dc47ce5b8fcfdb0c0e4454dbb77bbe73c6f4f9dbaa5", "generation_seconds_batch": 22.777452719048597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10530802607536316, "target_z": -0.2636718689279691, "ce": 6.260893821716309, "genuineness": 4.356721878051758, "identity": 0.363988995552063, "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/emo_Pain/aa971def5ae7062694d490a1f8f4d96e49b3fa0451ab580fed910361f41020f2.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "232e214702a131eeb0ecb2f47120fd3441063c8158055829b9887bb20806b33a", "generation_seconds_batch": 16.90793301607482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10643140971660614, "target_z": -0.2617187410943547, "ce": 6.4515180587768555, "genuineness": 4.080576419830322, "identity": 0.339613676071167, "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/emo_Pain/ab7a94299f968552a7b966fa3391724b652d8e20483888a7767196f29a7f4a69.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "abc2ad3e1358015c819bd2258561d1cd9729f5b28eff259d6dcdf6371b9dd9a5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b7741c9d395e9c241abb263ee39c712a58ce8299ea3659cc894070456d5bb742", "generation_seconds_batch": 21.722229861072265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1524900645017624, "target_z": -0.18164062945282264, "ce": 6.038498878479004, "genuineness": 4.821148872375488, "identity": 0.7756693363189697, "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/emo_Pain/abc2ad3e1358015c819bd2258561d1cd9729f5b28eff259d6dcdf6371b9dd9a5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "ac2fa28bb3f5534e376def22400e161ffd8e8c5db4d98a53052bce005210f25d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f26123f6b06faed11cf06fa7451df911d4164b860682cbc2ec3d5eae3d7f5972", "generation_seconds_batch": 18.554241056088358, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5715115666389465, "target_z": 0.5468750161920823, "ce": 5.959072113037109, "genuineness": 4.369274139404297, "identity": 0.395605206489563, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/ac2fa28bb3f5534e376def22400e161ffd8e8c5db4d98a53052bce005210f25d.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "88bece4d1e46ace1a2ce8ec9c6584051c66736efdde37856d253cce6ec33a772", "generation_seconds_batch": 16.72050570405554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 6.23706579208374, "genuineness": 4.432034492492676, "identity": 0.6287174820899963, "hypothesis": "Je bericht kwam nadat ik vertrokken was. Ik las ze twee keer en legde mijn telefoon weg. Zeg maar wanneer we kunnen praten.", "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/emo_Pain/adcc0121d6c11de6953f8243f84ad890832f75283c96342c1db246fddc443438.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "addc9186721381fdd6ea64de4e46e7cfc254ca4863f5d5c2cddb51e1f403e2df", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c9e495c9a7bceb388ee846beaca9dad0bc42ad9bad6d93f0d6ba503ed3006398", "generation_seconds_batch": 21.106439461000264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 6.308549880981445, "genuineness": 4.821148872375488, "identity": 0.3781312108039856, "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/emo_Pain/addc9186721381fdd6ea64de4e46e7cfc254ca4863f5d5c2cddb51e1f403e2df.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "ae2fdd9fb384bd1eee10c8290ed2eebdaf53636a56ba74416f12152b27d96a5e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "734709991eed46f2fdf454d87ec99fbc87b870eb46c1c0bd507fbf298b630872", "generation_seconds_batch": 22.603523959987797, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0633628368377686, "target_z": 3.14062494494692, "ce": 5.871703147888184, "genuineness": 4.15588903427124, "identity": 0.23363488912582397, "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/emo_Pain/ae2fdd9fb384bd1eee10c8290ed2eebdaf53636a56ba74416f12152b27d96a5e.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "c1f28a3f2796a28f63932f497aac07350a7224c76f9aef93a59fc1bdc5f3db77", "generation_seconds_batch": 17.73184676002711, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08733391761779785, "target_z": -0.29492186245113616, "ce": 6.332377910614014, "genuineness": 4.344170093536377, "identity": 0.31530293822288513, "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/emo_Pain/ae3130a5db36c2790a31e23135fde257111c56fa9c5b8061c1f3e51a8cb61258.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "aee78864461320cc3f5dccaf239928b59e11141b779e3b5a04416ede6bf6c684", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "61f69323a5470e1d97a11641f2d0de1534fcf08c9bbb549abd1fd51c45b66d6c", "generation_seconds_batch": 22.761081618024036, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2996531128883362, "target_z": 0.07421875404802059, "ce": 5.95112943649292, "genuineness": 4.7332844734191895, "identity": 0.6341410875320435, "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/emo_Pain/aee78864461320cc3f5dccaf239928b59e11141b779e3b5a04416ede6bf6c684.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": "b07b0e13e82405b7068f8e50072d976c570abe4b72d55040bd3913269b6e01b5", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "30f844f4edc1b07aa1957a8b57124dc983ccfd78768c4e4749c2c2d0d152ae43", "generation_seconds_batch": 17.39409685600549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7714735269546509, "target_z": 0.8945312006141488, "ce": 6.165581703186035, "genuineness": 4.20609712600708, "identity": 0.5853408575057983, "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/emo_Pain/b07b0e13e82405b7068f8e50072d976c570abe4b72d55040bd3913269b6e01b5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "b09fab71045a4b23b9ec3919105dea354c3cc430bf65a300d892dd46a985867b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c047e2ba1ecd46f65a8d15e9c66a8ca493cc13782c4a9a4ff3167a77eae3e6ac", "generation_seconds_batch": 23.74732047901489, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3951405882835388, "target_z": 0.2402343867392597, "ce": 5.557967185974121, "genuineness": 4.720732688903809, "identity": 0.7133948802947998, "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/emo_Pain/b09fab71045a4b23b9ec3919105dea354c3cc430bf65a300d892dd46a985867b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "b122e250a797081a1dab1f551c58218427b904a41861ac5633ea397f1e0a2a96", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "508099eaea02607b5e26cfbfea3bcf2ceba88a976cf9c83832c8123fdafc5725", "generation_seconds_batch": 24.8094939639559, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4881912469863892, "target_z": 2.14062494494692, "ce": 5.387199878692627, "genuineness": 4.293961524963379, "identity": 0.540942370891571, "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/emo_Pain/b122e250a797081a1dab1f551c58218427b904a41861ac5633ea397f1e0a2a96.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "b1302b8b237c31540d3e871450e267ef69777108ce336ffe6606188df9de86a2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e89b5553de153696dd477a1906c98ffb51597f58ce44074c576c97f972956e67", "generation_seconds_batch": 22.726454457035288, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.017684221267700195, "target_z": -0.4160156326912391, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.38456088304519653, "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/emo_Pain/b1302b8b237c31540d3e871450e267ef69777108ce336ffe6606188df9de86a2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "b1a4a667b40b13d7fe7ead942c6fc21361229b3b03bb26690f8a8c2fa539d564", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "b212738fd7f56f03664290dbcddf858e1e9bb5234c57911cab9afd5b6af468c7", "generation_seconds_batch": 22.834977486985736, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1601637601852417, "target_z": 1.5703125501954553, "ce": 6.3641486167907715, "genuineness": 4.406930446624756, "identity": 0.37659770250320435, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/b1a4a667b40b13d7fe7ead942c6fc21361229b3b03bb26690f8a8c2fa539d564.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": "b1b1d99cfcc69e1fc3078222919f4a0e3058f2706ca4c8c52a0567aa069e3459", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "adf2b683dd05edb841457083ec9fd5ec0e437d14706d9603c4dc5f2ab29a5de2", "generation_seconds_batch": 18.635635921964422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.039028480648994446, "target_z": -0.3789062556672288, "ce": 6.109982967376709, "genuineness": 4.419482707977295, "identity": 0.6268150806427002, "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/emo_Pain/b1b1d99cfcc69e1fc3078222919f4a0e3058f2706ca4c8c52a0567aa069e3459.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "b1ffbd179d8a060d068cd847dd13613b7fbd47c2765790fd1aee8276fe6417e0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cf60d09da7ec6d61c2d16a9a121cd0d734a333687763bfd0bc49a5b11d1eae11", "generation_seconds_batch": 22.579880341072567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07497671246528625, "target_z": -0.31640624271356294, "ce": 6.260893821716309, "genuineness": 4.582659721374512, "identity": 0.29711058735847473, "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/emo_Pain/b1ffbd179d8a060d068cd847dd13613b7fbd47c2765790fd1aee8276fe6417e0.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": "b206607d8c5bc5ca3e1c0074e0aecebef5cbf8c4e259ccd96665be405f46bae8", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2080d2951c94723c335a2e2ce33fef737b59f40bf404d13ed319ff692391bcc3", "generation_seconds_batch": 22.583702816977166, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.407307744026184, "target_z": 1.9999999481853366, "ce": 5.260117053985596, "genuineness": 3.961331605911255, "identity": 0.5566059350967407, "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/emo_Pain/b206607d8c5bc5ca3e1c0074e0aecebef5cbf8c4e259ccd96665be405f46bae8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "b20e78684c71262974e9389bd81b296c3f9b01608c69b98979de6d429ab1f690", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8597c4c400a3a7854859061299aa396bc7cf6b0a635e6484bb9f3cd379c8aecd", "generation_seconds_batch": 22.520148242940195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12215875089168549, "target_z": -0.23437500323841648, "ce": 6.070269584655762, "genuineness": 4.532451152801514, "identity": 0.6217053532600403, "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/emo_Pain/b20e78684c71262974e9389bd81b296c3f9b01608c69b98979de6d429ab1f690.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "b29a1daf0b3c9a78b1bcde29678eeebd6fae7777eb1c5d009f5ade9b6754dfbb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "89a32df3f56b32ad0403db9e552f554ce228b6ed5a387bc5a7d5eed138c8ce4b", "generation_seconds_batch": 21.64186190092005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.054755836725234985, "target_z": -0.35156249190395883, "ce": 5.974957466125488, "genuineness": 4.720732688903809, "identity": 0.47911304235458374, "hypothesis": "He encontrado la cajita. Estaba junto a. a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/b29a1daf0b3c9a78b1bcde29678eeebd6fae7777eb1c5d009f5ade9b6754dfbb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "b34850341bad2e4bbc0d07c46198d780964c35a7e72d31223a10c5eda5c946cc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ac730c564dacfcc8ccfb633b1f94d572fec8a2a3e35d86cd9f620536fa7ab873", "generation_seconds_batch": 16.470614130957983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6231871247291565, "target_z": 0.6367187410943547, "ce": 6.038498878479004, "genuineness": 4.419482707977295, "identity": 0.3644247055053711, "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/emo_Pain/b34850341bad2e4bbc0d07c46198d780964c35a7e72d31223a10c5eda5c946cc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "b35f5d603a6f78c4b12494e81c7940823a2c87ace1fe05309e10e431fb4b143e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dfed48e0febb89599b4ce83754d7d945858be265fde30cca3484cbb5daec7b3a", "generation_seconds_batch": 22.203985195956193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15024331212043762, "target_z": -0.1855468592127197, "ce": 6.141753673553467, "genuineness": 4.369274139404297, "identity": 0.692226231098175, "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/emo_Pain/b35f5d603a6f78c4b12494e81c7940823a2c87ace1fe05309e10e431fb4b143e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "b3a5f0dd4edc7a62af5a5392581de35a0fafc2faf4cf1671d99febf0e49028aa", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 1.0, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "eebf20950ed5b36359225b20393872535dc92e442c873ae032489692cc0b9097", "generation_seconds_batch": 25.72819890396204, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5894856452941895, "target_z": 0.5781249579005859, "ce": 5.657250881195068, "genuineness": 4.7081804275512695, "identity": 0.6757956147193909, "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/emo_Pain/b3a5f0dd4edc7a62af5a5392581de35a0fafc2faf4cf1671d99febf0e49028aa.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": "b3c9a8c7dcfd0f0545a61710451c9bbd3a51680709076d862c7161d6a59144df", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3f99f11d06fc201e661436bfcb69c075a96d6c12433f93c919aa4362566e8a5b", "generation_seconds_batch": 22.70014461502433, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7737202644348145, "target_z": 0.8984374044667142, "ce": 5.911416053771973, "genuineness": 4.406930446624756, "identity": 0.7125846743583679, "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/emo_Pain/b3c9a8c7dcfd0f0545a61710451c9bbd3a51680709076d862c7161d6a59144df.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "7784ba94eb53dc005ea06c44bf309a0d6949e9be995a60eb4347980c8a026302", "generation_seconds_batch": 16.269342013052665, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.109982967376709, "genuineness": 4.369274139404297, "identity": 0.7051156759262085, "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/emo_Pain/b429ad869b4f073af736e4e6510807542fc4f5324b0eb7f096c719d9002f4f71.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "b44da595583819abbb5ea9996755c80afd4308cf9eedbe5ae45fe77851071069", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "934d3eb007e0d7c44000eb0c47a316ce8906fd18b99e96631aa49e9ed5ff6e91", "generation_seconds_batch": 22.851109010982327, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11766523122787476, "target_z": -0.24218748866554235, "ce": 6.165581703186035, "genuineness": 4.507347106933594, "identity": 0.5342496037483215, "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/emo_Pain/b44da595583819abbb5ea9996755c80afd4308cf9eedbe5ae45fe77851071069.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "b46f6f408f1b2b8583e7ca71322d00f4e0db5acaf8782ef4aa798bf79e2ceac8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "827ce2dc9c9071f706a97aac4620b57582b1cc274fb38e84a4be2b4a71954567", "generation_seconds_batch": 27.145551898982376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2373054027557373, "target_z": -0.03417969822725456, "ce": 6.109982967376709, "genuineness": 4.645420074462891, "identity": 0.6614550352096558, "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/emo_Pain/b46f6f408f1b2b8583e7ca71322d00f4e0db5acaf8782ef4aa798bf79e2ceac8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "b4cf81764544b78f66e25b53e1f7fc24899dab56ce37cd4c3fb6a6fb7c121d66", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "214117e082d403744ace0d81a87f3572793e912dab6ad76c6bcef5b549ec2d53", "generation_seconds_batch": 22.627415099996142, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.046441555023193, "genuineness": 4.18099308013916, "identity": 0.2656141519546509, "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/emo_Pain/b4cf81764544b78f66e25b53e1f7fc24899dab56ce37cd4c3fb6a6fb7c121d66.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "b4cfb71d3b1e46edc69424270b2119dc8c2025f8e69ad40174112155ed5fc73d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cd8f16548b906537aa41a087b79a15fb2d78af56e5912acf0fca820c92af828c", "generation_seconds_batch": 18.77771246910561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07610009610652924, "target_z": -0.31445311487994854, "ce": 5.959072113037109, "genuineness": 4.68307638168335, "identity": 0.669022262096405, "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/emo_Pain/b4cfb71d3b1e46edc69424270b2119dc8c2025f8e69ad40174112155ed5fc73d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "b579986fcade814ee7c42fc5bb8650705bd13e97cbf7c57f4e379d9defe0988b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "16d342ed259dc5a777e22f9cac2cde322454cd9a3c2e52ae4c8c538f2bd35126", "generation_seconds_batch": 24.254576132982038, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.078346848487854, "target_z": -0.31054688512005146, "ce": 6.109982967376709, "genuineness": 4.846253395080566, "identity": 0.668677568435669, "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/emo_Pain/b579986fcade814ee7c42fc5bb8650705bd13e97cbf7c57f4e379d9defe0988b.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": "b5c27f23608342f5a803847dfde40ca14244bb74b0bcd9d96afb000b6063c735", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2bceeb705a34c6b3b0b3cb6a140d920824b812d28e5732f912f8b773e6ee2332", "generation_seconds_batch": 19.215147578972392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3097635507583618, "target_z": 0.09179687864321853, "ce": 5.9431867599487305, "genuineness": 4.394378662109375, "identity": 0.6116135716438293, "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/emo_Pain/b5c27f23608342f5a803847dfde40ca14244bb74b0bcd9d96afb000b6063c735.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "b68b8a0890b24243661fb1a1144dc6ec9c00d756da4fa72e6831312b61b48386", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5c8cff7090e198f6b4da9673167b8b2a820ab96de39618a807d5c7135b36edfb", "generation_seconds_batch": 22.949783375952393, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10081449151039124, "target_z": -0.27148438026242677, "ce": 6.252951145172119, "genuineness": 4.570107460021973, "identity": 0.4856187701225281, "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/emo_Pain/b68b8a0890b24243661fb1a1144dc6ec9c00d756da4fa72e6831312b61b48386.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "b6f8ecf3487e288dd2fd2e46fd5cb1e4e1ab331a0fd49d5f7044701fe22c48a2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "16cf2f12351d02c523d25985552764e283edb3bc6a4a891f64b6e892d0673883", "generation_seconds_batch": 23.408560162992217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6029662489891052, "target_z": 0.6015624919039588, "ce": 5.49045467376709, "genuineness": 4.381826400756836, "identity": 0.4099389314651489, "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/emo_Pain/b6f8ecf3487e288dd2fd2e46fd5cb1e4e1ab331a0fd49d5f7044701fe22c48a2.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": "b7628b835fea2426fd31b3780b0b03e45b6e03235d18164f35e0afb69d85a878", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d020ab6d97f8729c3be01cb8d3f2f3ec09524d5267930c4fd7782acba3aa6888", "generation_seconds_batch": 16.120575000997633, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1017478704452515, "target_z": 1.4687500064768328, "ce": 5.903473377227783, "genuineness": 3.9299514293670654, "identity": 0.07959636300802231, "hypothesis": "De weg gaat linksaf naar de brug. Loop verder. Tot aan dat 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/emo_Pain/b7628b835fea2426fd31b3780b0b03e45b6e03235d18164f35e0afb69d85a878.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "b784083811e37b4452dfca9accf40a000016385e5a5501d9d7ddbdf10797879a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "66050986862300a857dc3350de9047c6c12887b1f08791da799d8b1d67f725d9", "generation_seconds_batch": 21.9986300150631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7265383005142212, "target_z": 0.8164062945282264, "ce": 5.80021858215332, "genuineness": 4.444586753845215, "identity": 0.16939058899879456, "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/emo_Pain/b784083811e37b4452dfca9accf40a000016385e5a5501d9d7ddbdf10797879a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "b7fe4ec06ad08d4f03652a1488561052390958b0bbe16042af89c62062b55a82", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9d428cbcb857a23f50994af9665b310643b74c883c138cd020dd8932996878b6", "generation_seconds_batch": 22.738668750040233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11204831302165985, "target_z": -0.2519531278336144, "ce": 6.284721851348877, "genuineness": 4.632867813110352, "identity": 0.429995596408844, "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/emo_Pain/b7fe4ec06ad08d4f03652a1488561052390958b0bbe16042af89c62062b55a82.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "b8202f5094d6788b636b863ea2a91e586234d991c6497b1f14039a05207218fc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We can look inside together.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I found the little box.\n(Natural clear speech.) [3.8 seconds duration] It was beside the old window.\n(Natural clear speech.) [3.2 seconds duration] We can look inside together.", "seed": 1701, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "9ef98a7f1da4a9abb2ddc44e7a36431b02c608a0045c8234d99cf4830e49bc80", "generation_seconds_batch": 23.381265413947403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14069455862045288, "target_z": -0.20214843284477632, "ce": 6.268836498260498, "genuineness": 4.6705241203308105, "identity": 0.3378395438194275, "hypothesis": "I found the little box. It was beside the old window. We can look inside together.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/b8202f5094d6788b636b863ea2a91e586234d991c6497b1f14039a05207218fc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "b8ba15d0c43ceb4ca66d0df480ca68af164da922fd6c2e4fa8e6e1b4db792270", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1670d4bd7a50eac4c9e55483a0c82156855116d47e7589a5f96d0e801e97c03e", "generation_seconds_batch": 23.849411083036102, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.155670166015625, "target_z": 1.5624999352316706, "ce": 5.927301406860352, "genuineness": 4.808597087860107, "identity": 0.6195854544639587, "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/emo_Pain/b8ba15d0c43ceb4ca66d0df480ca68af164da922fd6c2e4fa8e6e1b4db792270.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": "b8be72527806a7715b9e4a340154dddacf04bbbd3345cbdc3fb8493229dcf5c5", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Call me when you have a moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] I wrote the number on the envelope.\n(Natural clear speech.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Natural clear speech.) [3.2 seconds duration] Call me when you have a moment.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9bf2793d7bb8637c0459e14d0437353226cfac2bae050a0ae5ebc5f5253fd8fc", "generation_seconds_batch": 17.459654210018925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09969110786914825, "target_z": -0.27343750809604117, "ce": 6.2767791748046875, "genuineness": 4.13078498840332, "identity": 0.5504117012023926, "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/emo_Pain/b8be72527806a7715b9e4a340154dddacf04bbbd3345cbdc3fb8493229dcf5c5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "b8c11b54a3b5d54963655c3091fd3d959889259182a3d1c99ae4080afd2a5b1b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "04352a81e57efa95a929c492bdd9bf711f763efd8cab4a15a5490942cef2f356", "generation_seconds_batch": 16.93266217398923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11598014831542969, "target_z": -0.24511719336962984, "ce": 6.165581703186035, "genuineness": 4.8588056564331055, "identity": 0.7404171228408813, "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/emo_Pain/b8c11b54a3b5d54963655c3091fd3d959889259182a3d1c99ae4080afd2a5b1b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "b960089b8ab514800896b2a147e868b97ced606579bad273d7e01672f5f50bc5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1701, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7dd43bc9f0ac674218a4c023ddca177c6f400c844d3a22339a42ac704648de04", "generation_seconds_batch": 23.586297124042176, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8926087617874146, "target_z": 2.843749928754838, "ce": 5.784333229064941, "genuineness": 4.086852550506592, "identity": 0.5417187213897705, "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/emo_Pain/b960089b8ab514800896b2a147e868b97ced606579bad273d7e01672f5f50bc5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "b9debaafb0ff961caaf63b1769779d80ad5b67ee368cb957188a46a5f58bf793", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0d66d29bd60b1cbbbd3e3b5091cfde6a4f679174b573d8b4c446d53206821425", "generation_seconds_batch": 16.93434958695434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.19967210292816162, "target_z": -0.09960938997767617, "ce": 5.895530700683594, "genuineness": 4.7081804275512695, "identity": 0.7274283170700073, "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/emo_Pain/b9debaafb0ff961caaf63b1769779d80ad5b67ee368cb957188a46a5f58bf793.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "b9e58784925c1ec65b0b199ee808edb605eda4ea6b6d99894886a432447e8fe8", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ba6b084c02b73336cf1b81f4b09f550aaf817d09f11ba39d7b3321b4c63bd92a", "generation_seconds_batch": 18.01512525510043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 6.221180438995361, "genuineness": 4.494794845581055, "identity": 0.6094086170196533, "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/emo_Pain/b9e58784925c1ec65b0b199ee808edb605eda4ea6b6d99894886a432447e8fe8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "ba1a5e8f301393c2de65f6ad7cd53bb8c43ceb65e62b11fad40ee2ed894fe75d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "eb693d76684eec7cfae0d9e68e6d348ce252ebb1f2a236f2014493971073c251", "generation_seconds_batch": 18.43492199399043, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6859065294265747, "target_z": 2.484375029145748, "ce": 5.855817794799805, "genuineness": 4.331617832183838, "identity": 0.5407261848449707, "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/emo_Pain/ba1a5e8f301393c2de65f6ad7cd53bb8c43ceb65e62b11fad40ee2ed894fe75d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "ba25c7fa07508389e82d1376b1aedd2c32760051dc34564367ad1ffe54867234", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0eae1d1eb3e29ad8f40c246723ba59ba911c7f1aee66f1c374d21626da37b867", "generation_seconds_batch": 23.23650970601011, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.117925643920898, "genuineness": 4.645420074462891, "identity": 0.8205353021621704, "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/emo_Pain/ba25c7fa07508389e82d1376b1aedd2c32760051dc34564367ad1ffe54867234.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "4e48e64cc237ac6e962f7acbdb3337e4b1580be85e4992793f0a990e725ad5e2", "generation_seconds_batch": 16.531670378055423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.078212261199951, "genuineness": 4.595211505889893, "identity": 0.6883010864257812, "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/emo_Pain/ba70e7e208a6b15a9236ccd5a69e0e5fd403df2a937202d690effa2cdc45d9f2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "baa9c2724dbba515d03bac6cea2b0875728bef28846ebb1004bb1c686dc9c1f6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fa061de8488295e69a8ff67a3f5f8e029abcbf83acac2833d1110fd9a3889a5e", "generation_seconds_batch": 21.86315690097399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.047825574874878, "target_z": 1.3750000777219953, "ce": 5.557967185974121, "genuineness": 4.595211505889893, "identity": 0.7018758058547974, "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/emo_Pain/baa9c2724dbba515d03bac6cea2b0875728bef28846ebb1004bb1c686dc9c1f6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "bad8a9870d73128314107008d3b26c0473557d182a43c74be52f86ddf921d3e6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "06e74f79b800aac64797af3f3e3ab70810b1f9dcbf6254f90bf70d174abfd164", "generation_seconds_batch": 27.67689431400504, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07048317790031433, "target_z": -0.3242187540480206, "ce": 6.348263263702393, "genuineness": 4.6203155517578125, "identity": 0.5148569941520691, "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/emo_Pain/bad8a9870d73128314107008d3b26c0473557d182a43c74be52f86ddf921d3e6.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": "bb744e540a73db7605f9c4a4e4f704178af2bdd10908659a198d91dc50768ac1", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "f9872ab68ceca8d88c9e8fb71b5dab4e3f792f0cc39ea748ef8070de63708693", "generation_seconds_batch": 17.480173617019318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13114580512046814, "target_z": -0.21875000647683293, "ce": 5.8875885009765625, "genuineness": 4.268857479095459, "identity": 0.55681973695755, "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": 0.03999999999999915, "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/emo_Pain/bb744e540a73db7605f9c4a4e4f704178af2bdd10908659a198d91dc50768ac1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "bc4bc19e4814b54628b087407b5fad0f1803a5a17502d119e3feb12a419958ff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "93bc4e6384af710393fbc4b6b2de25edf990f3f4718ed50a04fca62edf27be2b", "generation_seconds_batch": 16.79077940399293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5984727144241333, "target_z": 0.5937499805695012, "ce": 5.903473377227783, "genuineness": 4.118232727050781, "identity": 0.5466176271438599, "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/emo_Pain/bc4bc19e4814b54628b087407b5fad0f1803a5a17502d119e3feb12a419958ff.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "bd092fbef4344f434d30e1ba3ccb69e23bb04f5888467c2b282a401010732133", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c7744141be8881e2985cce10b1b3e636e510e4fd1193f4d2b3d353dfe9b6467e", "generation_seconds_batch": 23.381265413947403, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1940551996231079, "target_z": -0.10937500323841647, "ce": 6.0146708488464355, "genuineness": 4.482243061065674, "identity": 0.5708637833595276, "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/emo_Pain/bd092fbef4344f434d30e1ba3ccb69e23bb04f5888467c2b282a401010732133.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "bd6214b586c15ab8584f8d6c381ad9eac1897a00df39a9ea62b378a9518fe4eb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Natural clear speech.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Natural clear speech.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "seed": 1702, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9f8fe9e3033ba941daa25dc3fb55f69a3fbf092abe1ff32826cfa014e9b3c0e9", "generation_seconds_batch": 26.7373935719952, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08396376669406891, "target_z": -0.3007812459519794, "ce": 6.23706579208374, "genuineness": 4.7834930419921875, "identity": 0.6637770533561707, "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/emo_Pain/bd6214b586c15ab8584f8d6c381ad9eac1897a00df39a9ea62b378a9518fe4eb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "bd63d11100d8f3a5f0c9174a165f0bcde5f78ce0a1b6f049a9a152077a2646ff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 1.25, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "911418ea1e250b800940255ae0eb7a83bd1bc8f1a459c63e5ef3f819c456b07d", "generation_seconds_batch": 22.221234102966264, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.133954256772995, "target_z": -0.2138671998464628, "ce": 6.0305562019348145, "genuineness": 4.6705241203308105, "identity": 0.644214391708374, "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/emo_Pain/bd63d11100d8f3a5f0c9174a165f0bcde5f78ce0a1b6f049a9a152077a2646ff.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "bda1672d970dcca56e62b94386e28b58b1e85783806632544cea434ebe1b49e5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3e25a86d84bc6f9f7c4fc1249df6913bc69e5d2a5db45400fa319f5db96a69b2", "generation_seconds_batch": 19.283897335990332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.753499448299408, "target_z": 0.8632812589056453, "ce": 5.8875885009765625, "genuineness": 4.65797233581543, "identity": 0.7425290942192078, "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/emo_Pain/bda1672d970dcca56e62b94386e28b58b1e85783806632544cea434ebe1b49e5.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": "bdb5428704855e27884f60a0db8e92d7ac80fef6100462555b410ed3042f4ebb", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Natural clear speech.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Natural clear speech.) [3.2 seconds duration] Nimm sie mit, wenn du gehst.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "7ddb369619cb9fc8ceab84963b699020aafb8e9f28660ff0e0fa24145b45e504", "generation_seconds_batch": 19.075753710931167, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3058317005634308, "target_z": 0.08496091827190222, "ce": 5.903473377227783, "genuineness": 4.7081804275512695, "identity": 0.642943263053894, "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/emo_Pain/bdb5428704855e27884f60a0db8e92d7ac80fef6100462555b410ed3042f4ebb.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": "be0d393690777ed965783b6d6fac422d39478f519a62c519c2952c7e780d34cb", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2eb3d7afc085ff19bb5a943dd61d34d40d3b2481c52ac9a48570351c0807319b", "generation_seconds_batch": 16.941421539057046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.445692777633667, "target_z": 0.3281250097152494, "ce": 5.4944257736206055, "genuineness": 4.2249250411987305, "identity": 0.6085540652275085, "hypothesis": "La carretera gira a la izquierda tras el puente. Sigue hasta la pequeña puerta de mi 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/emo_Pain/be0d393690777ed965783b6d6fac422d39478f519a62c519c2952c7e780d34cb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "be2c50cda761e67a5dca6b7983f17ed046eda121721babbc46837b8a44ab98bd", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "204284463f8b2fd7aae4964b1369f4efc1829adddf457476d0dad44c8795f637", "generation_seconds_batch": 24.02667941409163, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.133810997009277, "genuineness": 4.6705241203308105, "identity": 0.6757123470306396, "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/emo_Pain/be2c50cda761e67a5dca6b7983f17ed046eda121721babbc46837b8a44ab98bd.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "be44efb33c9a449676ed7c9320e1d705f7a39d0334fd99297dedf8a6cf720afe", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "80c6c113903959033d21b1d02fa36ecb4eeab97aea75158369ce7f295186324c", "generation_seconds_batch": 23.739762646961026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.31875061988830566, "target_z": 0.10742190131213382, "ce": 5.847874641418457, "genuineness": 4.607763767242432, "identity": 0.6547940373420715, "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/emo_Pain/be44efb33c9a449676ed7c9320e1d705f7a39d0334fd99297dedf8a6cf720afe.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "becc856a5921f0fab6a97b7128740761fd8f83a213bbca41544a0fd76fc25fde", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e174be85877b2121f888842c208f4e5391fac5fad0e902d8b89053107ae73c9d", "generation_seconds_batch": 24.02672382502351, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 6.046441555023193, "genuineness": 4.532451152801514, "identity": 0.6898921728134155, "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/emo_Pain/becc856a5921f0fab6a97b7128740761fd8f83a213bbca41544a0fd76fc25fde.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": "bf7f2f6d6d5d0ee6f515ea2ce879b57f94967fd0fb3d0263beefae0b06e86e7b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Straße führt hinter der Brücke nach links.\n(Natural clear speech.) [3.8 seconds duration] Geh weiter... bis zu dem kleinen Holztor.\n(Natural clear speech.) [3.2 seconds duration] Ich warte auf der anderen Seite.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "0c0949c62303719e76b8742872772f7218d5e803199f825faab91ee0951d5190", "generation_seconds_batch": 16.938118551974185, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 5.927301406860352, "genuineness": 4.494794845581055, "identity": 0.7913281321525574, "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/emo_Pain/bf7f2f6d6d5d0ee6f515ea2ce879b57f94967fd0fb3d0263beefae0b06e86e7b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "c047b48a820dc9dd39611a2d110c8b2eb8924c5de0a10b0035f0b75e935e78e4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2d750c5f8c43858fda9dd003a23acb432e83e0f1588dab8b4d8b4fd75e183f65", "generation_seconds_batch": 22.945940076955594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4501863121986389, "target_z": 0.33593752104970703, "ce": 6.205295085906982, "genuineness": 4.319066047668457, "identity": 0.44931548833847046, "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/emo_Pain/c047b48a820dc9dd39611a2d110c8b2eb8924c5de0a10b0035f0b75e935e78e4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "c1086962387fdbd94617df971499128a250f25e19ac7685ccd8a1f5e29f7d2ed", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "79e40b31e5450318b56107bfb694c977737db676dc64089d1b443d2c8b3238e9", "generation_seconds_batch": 17.632916691945866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28659379482269287, "target_z": 0.05151367212800129, "ce": 5.526196479797363, "genuineness": 4.519899368286133, "identity": 0.5051943063735962, "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/emo_Pain/c1086962387fdbd94617df971499128a250f25e19ac7685ccd8a1f5e29f7d2ed.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": "c131130086af06d00417fd0f3a8f109c2f7eb2494c0366e10309d08117d6eb65", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I wrote the number on the envelope.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Read it carefully... and check the last two digits.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dd632fa37659b004c72e4b3ee04a196284c597f84111c10096522118820d90c6", "generation_seconds_batch": 17.296011709957384, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1889999806880951, "target_z": -0.11816406553601544, "ce": 6.268836498260498, "genuineness": 4.545003414154053, "identity": 0.5316469669342041, "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/emo_Pain/c131130086af06d00417fd0f3a8f109c2f7eb2494c0366e10309d08117d6eb65.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "c1b1b3d422426e4d32e71a1d8c48de53b687a845bd686d0be793e1d7d153b28f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "236632d04108c397f502835620122aa8f1c67d7f6cc470b7d80bcec8c3785117", "generation_seconds_batch": 30.399186848080717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1148567646741867, "target_z": -0.24707032120324426, "ce": 6.0305562019348145, "genuineness": 4.65797233581543, "identity": 0.5818396806716919, "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/emo_Pain/c1b1b3d422426e4d32e71a1d8c48de53b687a845bd686d0be793e1d7d153b28f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "c1b3f976db0fa85c08380e2c3d07b987b8f09aaa82670c5dce7ec838b8f5f7c4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "8cec11267599af1f61a0b6182df092718b6a02f1cf5b51854769ed65acb497a1", "generation_seconds_batch": 22.182554475031793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6613820791244507, "target_z": 0.7031249319932541, "ce": 6.006728172302246, "genuineness": 4.055472373962402, "identity": 0.7628954648971558, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/c1b3f976db0fa85c08380e2c3d07b987b8f09aaa82670c5dce7ec838b8f5f7c4.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": "c1b459f4da1e8f779315f78d88c0eedc5e6e01c98f6aa868c2a266d1f9cd748f", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "87f6753bef50043f5721c7d71d31afa34978b7d6d2b411b96de42d7f2167f934", "generation_seconds_batch": 23.666788005037233, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.1020402908325195, "genuineness": 4.607763767242432, "identity": 0.801036536693573, "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/emo_Pain/c1b459f4da1e8f779315f78d88c0eedc5e6e01c98f6aa868c2a266d1f9cd748f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "c1dae0bebc116bb2be5b5c79a8e337fcbfa336efbc19e82cb7841a8f8f3bbc74", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e1969c2411347614a6299bbd5dc8c6a9693a7293cb713ec4aa7545bea9674d71", "generation_seconds_batch": 23.261758264969103, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16709403693675995, "target_z": -0.15624999352316707, "ce": 5.998785495758057, "genuineness": 4.808597087860107, "identity": 0.5347711443901062, "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/emo_Pain/c1dae0bebc116bb2be5b5c79a8e337fcbfa336efbc19e82cb7841a8f8f3bbc74.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "c1e9ac833f161fc0586aba0cfc7c83d380aaaa6310bab973a932b7ce9de4609e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "acd604744d3e9ab13d913115150ce691ffd38dc50e1b764b1df1980c8d585252", "generation_seconds_batch": 25.888174471096136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.037905097007751465, "target_z": -0.38085938350084325, "ce": 6.133810997009277, "genuineness": 4.519899368286133, "identity": 0.5557889342308044, "hypothesis": "Ik heb het klein een doosje gevonden, het stond naast het oude raam. We kunnen er 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/emo_Pain/c1e9ac833f161fc0586aba0cfc7c83d380aaaa6310bab973a932b7ce9de4609e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "c1fd16f61ded60e6f4df12edf5b96d2bdedd5de49e3df2bd1ab72ba0a5ddff10", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6d722314ade811364ba39b83ead233a18feeac9171519821687bb746a8f4b310", "generation_seconds_batch": 16.78667964693159, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08059361577033997, "target_z": -0.30664062945282267, "ce": 6.054384231567383, "genuineness": 4.595211505889893, "identity": 0.5697448253631592, "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/emo_Pain/c1fd16f61ded60e6f4df12edf5b96d2bdedd5de49e3df2bd1ab72ba0a5ddff10.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": "c21f1e4f0af1858baa53608442ec0709b2c3745031ba1d05bb8a26f10296ed66", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Natural clear speech.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand nous pourrons parler.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "674fc2c4f1e399ba535300927b0b17bd1c1715c413637c6a11d10da472d09460", "generation_seconds_batch": 16.672387900063768, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.245540738105774, "target_z": 1.7187499546621694, "ce": 6.125868320465088, "genuineness": 4.7081804275512695, "identity": 0.7445296049118042, "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/emo_Pain/c21f1e4f0af1858baa53608442ec0709b2c3745031ba1d05bb8a26f10296ed66.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "c23c0e86395bb5e94763730678af4974b360d515fe7688485e76c14b04fa3a8d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5c16efd8184abf4d62a78accbbf8eb92bc5f36cfaa3eb3d5ff7d78552793e8d9", "generation_seconds_batch": 18.30679852794856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4432560205459595, "target_z": 2.0625000388609975, "ce": 5.808161735534668, "genuineness": 4.457139015197754, "identity": 0.5658776760101318, "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/emo_Pain/c23c0e86395bb5e94763730678af4974b360d515fe7688485e76c14b04fa3a8d.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": "c2470e8c9d8954c7e42198cd50ce91027c6585adc89d45ad322d683bdada674b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Natural clear speech.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer we kunnen praten.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a1a87471f60ac26b55dc40fd17049fd6a63ff1326be4a7aabb51362cd49c7c0d", "generation_seconds_batch": 17.114094709977508, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41199129819869995, "target_z": 0.2695312265214806, "ce": 6.0146708488464355, "genuineness": 4.419482707977295, "identity": 0.5292961597442627, "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/emo_Pain/c2470e8c9d8954c7e42198cd50ce91027c6585adc89d45ad322d683bdada674b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "c32384d62bc0e99fe4471a81bbd44326a66e194beedeff04f56e8021ab7c8f38", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9f0a53f125333377d6092e2a3accdcaad1dbfc42c46da47a12a9ead35683ff13", "generation_seconds_batch": 23.01767224597279, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.23706579208374, "genuineness": 4.6705241203308105, "identity": 0.504358172416687, "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/emo_Pain/c32384d62bc0e99fe4471a81bbd44326a66e194beedeff04f56e8021ab7c8f38.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "c35d252e82d559140bd0953c608dfdf4febefd8dd21c13780ad52ee7fd69b8fb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d158d27a36e3568148f27aab779a249ae3f3a19a7da6c62297fd3c2e773d3125", "generation_seconds_batch": 18.860914860037155, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4702171087265015, "target_z": 2.1093748996090893, "ce": 5.561938762664795, "genuineness": 4.118232727050781, "identity": 0.5256184339523315, "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/emo_Pain/c35d252e82d559140bd0953c608dfdf4febefd8dd21c13780ad52ee7fd69b8fb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "c35ebe8fde90fdf5ebc5499ee3f77f2246f9a88f3340e73b0fc1e079df40646e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "23d2b674596054d8a3b9d86cf09c119eec4ef0ece1825ac98ee439cde9e0e6b4", "generation_seconds_batch": 22.601961210020818, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 6.0305562019348145, "genuineness": 4.7458367347717285, "identity": 0.5050824880599976, "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/emo_Pain/c35ebe8fde90fdf5ebc5499ee3f77f2246f9a88f3340e73b0fc1e079df40646e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "c36e7c044319eac43e9a1c9e36e396c936b15a691ef28021cda81170032f8636", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4d84f07c2f969adba5583fa41aac0e03f9ce5fc6c6368b4c44606dc5629c3063", "generation_seconds_batch": 23.32275474607013, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.395919322967529, "genuineness": 4.545003414154053, "identity": 0.6046174764633179, "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/emo_Pain/c36e7c044319eac43e9a1c9e36e396c936b15a691ef28021cda81170032f8636.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "04ccbbdc8384397074d99b887be95d990cf82b0c7c0f748e45a9c489ccba7265", "generation_seconds_batch": 16.65194062003866, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.109982967376709, "genuineness": 4.582659721374512, "identity": 0.6679816246032715, "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/emo_Pain/c3977db0634952a7f9e4083e6135fbbc87199d34c7cb8c3338515d9ae38c18f4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "c3993072f7038ac355d6f6006f2c790d1a1cae7164a76c0a20830d8a52ec2501", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "25554d9d6e1e9693f1987536d096c8370860edaa4ef6a8e8a680ed85b284a0a4", "generation_seconds_batch": 22.965865625068545, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.40019580721855164, "target_z": 0.24902344903685866, "ce": 6.197352409362793, "genuineness": 4.7332844734191895, "identity": 0.45341262221336365, "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/emo_Pain/c3993072f7038ac355d6f6006f2c790d1a1cae7164a76c0a20830d8a52ec2501.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "0a56f6a4b5cf1ec9e4357c3d37f1dc1e3600bfbc23f39ba541dce2fecef8957a", "generation_seconds_batch": 16.847944028908387, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.039028480648994446, "target_z": -0.3789062556672288, "ce": 6.260893821716309, "genuineness": 4.457139015197754, "identity": 0.6677235960960388, "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/emo_Pain/c493619abd8899c9b3ce96f2a00a6bcdd900e874ef21436707b986365a8437ee.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": "c4a5bdef4acf981b8ec19c864635172cc314e3411df2fccf16b032854adeecd6", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "d29bbbb3ca0c2b46445ab3202676e2f57a5e7b6de8bf1e8d515e34e17146ac80", "generation_seconds_batch": 17.39409685600549, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9265003204345703, "target_z": 1.16406258257962, "ce": 6.252951145172119, "genuineness": 4.3127899169921875, "identity": 0.6715799570083618, "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/emo_Pain/c4a5bdef4acf981b8ec19c864635172cc314e3411df2fccf16b032854adeecd6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "c4c5d1ef5d6279c4fe04b12fd238d01a5515e9617373d1bd9a123974fdd243be", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "46a81e7273489ffa9dcad240927150a60cc9d2a26daa76b2f8fc8aae0edafc97", "generation_seconds_batch": 16.81523526192177, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21146762371063232, "target_z": -0.07910156067839073, "ce": 6.006728172302246, "genuineness": 4.118232727050781, "identity": 0.4244195818901062, "hypothesis": "I found a 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/emo_Pain/c4c5d1ef5d6279c4fe04b12fd238d01a5515e9617373d1bd9a123974fdd243be.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": "c4e61c5e1a5d768e18abf5d96fae7b75ef1f51646992b5b6d1ead9e20a7b208d", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "60211132f89fda1509e6686e2fd77de526406b001daf19c41bc37bf70198d4f0", "generation_seconds_batch": 15.773294360027649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0208643674850464, "target_z": 1.3281250097152495, "ce": 5.95112943649292, "genuineness": 4.406930446624756, "identity": 0.6579309701919556, "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/emo_Pain/c4e61c5e1a5d768e18abf5d96fae7b75ef1f51646992b5b6d1ead9e20a7b208d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "c5dff61e41b48d8a3d259157ac894f09b179e6957d4c58ed13ed8aa0f8576783", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5a9d3e8345f9ad89fae48ffe3eef30641e53d9260299e4ee3858b03b329eaf14", "generation_seconds_batch": 22.363918137038127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32099735736846924, "target_z": 0.11132810516469913, "ce": 5.824047088623047, "genuineness": 4.607763767242432, "identity": 0.7624785900115967, "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/emo_Pain/c5dff61e41b48d8a3d259157ac894f09b179e6957d4c58ed13ed8aa0f8576783.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "c640ca7fe9c953ae35e6e9cd4c7e72c01f3b9b24df3997c7e590c9cbd82371ef", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ac5657320a6e979da45b309221a68a3662a368c794f076b1cd850fb392395308", "generation_seconds_batch": 22.67614739399869, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23365440964698792, "target_z": -0.04052735720966845, "ce": 5.911416053771973, "genuineness": 4.7834930419921875, "identity": 0.6748862862586975, "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/emo_Pain/c640ca7fe9c953ae35e6e9cd4c7e72c01f3b9b24df3997c7e590c9cbd82371ef.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "c73ad58586c2505a39c12c9f07f4b3249c2919a8189c6449913f5a93294b9d4e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bdaea636e3f6975a3049236c1b4cd95f9614b8ad5ce48bfa192081cafe280481", "generation_seconds_batch": 22.13547286496032, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.025547906756401062, "target_z": -0.40234373785593824, "ce": 6.078212261199951, "genuineness": 4.532451152801514, "identity": 0.2756211757659912, "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/emo_Pain/c73ad58586c2505a39c12c9f07f4b3249c2919a8189c6449913f5a93294b9d4e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "c794069211f4ab2d64b2e178ddc7b93c7e01f15196dc61574d0a835aea18a6d1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "62f1fe09b2b1b6387f6c5ea256938406fc44e8849d75b35347dbe3edb5625dbb", "generation_seconds_batch": 17.256684998050332, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.21399523317813873, "target_z": -0.07470702952959125, "ce": 5.998785495758057, "genuineness": 4.595211505889893, "identity": 0.5800684690475464, "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/emo_Pain/c794069211f4ab2d64b2e178ddc7b93c7e01f15196dc61574d0a835aea18a6d1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "c7ad9f3b7109b4cdc6c140da14d46dd9b2c654691078bcf4f023c07fec62fa09", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8d87179c165c2fdd0244586ee286e7222cb2fcab780de39e3b154d989bb9b288", "generation_seconds_batch": 21.86315690097399, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7804605960845947, "target_z": 0.9101562232830641, "ce": 5.903473377227783, "genuineness": 4.65797233581543, "identity": 0.5982760787010193, "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/emo_Pain/c7ad9f3b7109b4cdc6c140da14d46dd9b2c654691078bcf4f023c07fec62fa09.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "c7da0fc7748b878665ac99a455f01b8e1bd61074b12327700aff46c299e8e15d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1701, "scale": 1.0, "cfg": 4.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "88d7204f88bc9340b7cd07fbe1002013bb2521ee8b0f740df4271e35716301b7", "generation_seconds_batch": 22.94222763599828, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8566606044769287, "target_z": 2.7812500453378304, "ce": 6.141753673553467, "genuineness": 4.15588903427124, "identity": 0.39486974477767944, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/c7da0fc7748b878665ac99a455f01b8e1bd61074b12327700aff46c299e8e15d.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": "c7f3018e860343f067ea5d19a5ddb6fa7ba288507fe73f79095e19173547a43c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2527719403688508e419f99d5c9870f9bd64cc30880b07579276d99017d9860f", "generation_seconds_batch": 21.618244612007402, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1328308880329132, "target_z": -0.21582030177274544, "ce": 6.1020402908325195, "genuineness": 4.21237325668335, "identity": 0.2312505841255188, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekkast. We zetten het na de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/c7f3018e860343f067ea5d19a5ddb6fa7ba288507fe73f79095e19173547a43c.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "74dda75af7772fa57c4b469466b2fb541092a49779eeddfdb0249bf7531b4daa", "generation_seconds_batch": 16.72050570405554, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": -0.028374433517456055, "target_z": -0.4960937443327712, "ce": 6.086154937744141, "genuineness": 4.306513786315918, "identity": 0.8579561710357666, "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": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/c82131bfaf89850d394bf3a21117deb94436b65306054c3e1cbac0c65e2f913d.mp3" }, { "id": "heldout-en-3", "language": "en", "sentences": [ "Someone moved the little blue chair.", "It used to stand here... beside the tall bookshelf.", "We can put it back after lunch." ], "text": "Someone moved the little blue chair. It used to stand here... beside the tall bookshelf. We can put it back after lunch.", "sentence_seconds": [ 3.2, 3.8, 3.2 ], "seconds": 10.2, "frames": 128, "max_frames": 188, "reference": true, "task_id": "heldout-en-3", "key": "c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d", "bucket": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "e2b618fc4baeb75de21906691935681b6cb22b91cce3b31d41ca090966668c70", "generation_seconds_batch": 17.42886963998899, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.05700258910655975, "target_z": -0.34765626214406176, "ce": 6.3641486167907715, "genuineness": 4.356721878051758, "identity": 0.3251006007194519, "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/emo_Pain/c8907269f81c0a33f09ba3b9d9f76b6acb2a8250e6881dbc11786cea01719e9d.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": "c8fcc4e2670d79b69c38b7b9d43c8c24a6fd45af527811b4a5191471131e5c4b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "efdfd33d477b72c0e3343b50fa39d1cfcbed76ddb0f451b67de684b457d01baf", "generation_seconds_batch": 17.480173617019318, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15473683178424835, "target_z": -0.1777343737855938, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.722697377204895, "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": 0.03999999999999915, "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/emo_Pain/c8fcc4e2670d79b69c38b7b9d43c8c24a6fd45af527811b4a5191471131e5c4b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "c90a76e42f489a3a996db026bf3046b4e22303e35acad62c9864696180e61fa5", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "18ae5750f4e73d4e4a8078145d7a124f60302c1827fe871185d277643582721b", "generation_seconds_batch": 16.47908243793063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04913891851902008, "target_z": -0.3613281310720309, "ce": 6.117925643920898, "genuineness": 4.482243061065674, "identity": 0.8257033824920654, "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/emo_Pain/c90a76e42f489a3a996db026bf3046b4e22303e35acad62c9864696180e61fa5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "c99fcfd874beb9425504cecb13bc2d92c45f8d2ac858158a3b6129e1bfd5cd04", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4d3384dc5a7c4828b629ef734fb9dfda7041e1250d41dc5a651f1ebcfb7a15a5", "generation_seconds_batch": 16.333416844019666, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3288610577583313, "target_z": 0.12500002590733175, "ce": 5.847874641418457, "genuineness": 4.833701133728027, "identity": 0.31348657608032227, "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/emo_Pain/c99fcfd874beb9425504cecb13bc2d92c45f8d2ac858158a3b6129e1bfd5cd04.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "c9a46a3ec0b384b9d890c6f31d95af7b299d4492ec539fc118c95965b5b2fcdb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "ec3095f2b4f1623c80357b85f11957bd70a56b945b6497223e2b1f263bbd3f1b", "generation_seconds_batch": 21.69359755294863, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.6859065294265747, "target_z": 2.484375029145748, "ce": 5.7406487464904785, "genuineness": 4.331617832183838, "identity": 0.46933820843696594, "hypothesis": "De hur staat nog open. Ik heb je jas op de stoel gelegd. Wanneer je klaar bent.", "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/emo_Pain/c9a46a3ec0b384b9d890c6f31d95af7b299d4492ec539fc118c95965b5b2fcdb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "c9a8fcc0e8b7ce55816d2176f1a642d67423ba02fd1a6405553ba15ae564c4c7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "57077094019dac919fc0305ecbd37cef026b9791b4dfe14f981e784e04bb474b", "generation_seconds_batch": 22.203985195956193, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4591733515262604, "target_z": 0.35156249190395883, "ce": 6.054384231567383, "genuineness": 4.507347106933594, "identity": 0.6602885723114014, "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/emo_Pain/c9a8fcc0e8b7ce55816d2176f1a642d67423ba02fd1a6405553ba15ae564c4c7.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "c9b2af705e2ab82b04bcc9d497246154129cdacfe974f8168cb74b71e8bcc60b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "338c6c8dce3715da0d65e240286876812126c0c00733bd7bd77f86991a645e4e", "generation_seconds_batch": 18.232507715001702, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.02217775583267212, "target_z": -0.4082031213567815, "ce": 6.062326908111572, "genuineness": 4.557555198669434, "identity": 0.5934070348739624, "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/emo_Pain/c9b2af705e2ab82b04bcc9d497246154129cdacfe974f8168cb74b71e8bcc60b.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": "c9dc4f0345b182e6b940cddf7c2fee0ea71cc0f8e30d6318d0fb948e772e8964", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Natural clear speech.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Natural clear speech.) [3.2 seconds duration] Bel me wanneer je even tijd hebt.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1fec18add5a9f273b3c3fa3cdd6f64189439eaf77d7bac9b8cd4572213c0dbeb", "generation_seconds_batch": 18.635635921964422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 6.157639026641846, "genuineness": 4.494794845581055, "identity": 0.6233820915222168, "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/emo_Pain/c9dc4f0345b182e6b940cddf7c2fee0ea71cc0f8e30d6318d0fb948e772e8964.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "ca39da1e1d4f3dd4706508615615114040102dc655aa3705e64a80e6215ba779", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "50eed3a17eaf6f5437f2e4b3a923c0f00841191196febb223084ce9a3c356b3e", "generation_seconds_batch": 24.395085487980396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3962639570236206, "target_z": 0.24218748866554235, "ce": 5.982900142669678, "genuineness": 4.7834930419921875, "identity": 0.6531578898429871, "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/emo_Pain/ca39da1e1d4f3dd4706508615615114040102dc655aa3705e64a80e6215ba779.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "ca6fd2a662bedbc6842a2417877d9c8655ac06227c03afbc1e28193d1db84a47", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ddca503a7b6eaf21a3d43abed905f5c744bde7aea32c66e0ef101664689ca8a5", "generation_seconds_batch": 22.09440491499845, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07610009610652924, "target_z": -0.31445311487994854, "ce": 6.205295085906982, "genuineness": 4.720732688903809, "identity": 0.6389849185943604, "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/emo_Pain/ca6fd2a662bedbc6842a2417877d9c8655ac06227c03afbc1e28193d1db84a47.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "cac7f85be4ea68b0590b3a5c9acc33f78a99d54c516618d8097b9f63652838bb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b715ac6b0022105097451b2395ae91d001cde8f49e9cc8e8d6433269dfc0d83d", "generation_seconds_batch": 17.04402014799416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14462639391422272, "target_z": -0.19531249838079176, "ce": 6.070269584655762, "genuineness": 4.256305694580078, "identity": 0.6356716156005859, "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/emo_Pain/cac7f85be4ea68b0590b3a5c9acc33f78a99d54c516618d8097b9f63652838bb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "cbd57885898dcb7319c3c990916fb69d927f1755efbf2d40c8d97480bcc093b5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2fc6ddbe016fe40bc0e865acfbbb590de45e7a2358119a0b35aced3a1634916b", "generation_seconds_batch": 23.408560162992217, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.497178316116333, "target_z": 2.1562499676158353, "ce": 5.1628193855285645, "genuineness": 4.055472373962402, "identity": 0.47084879875183105, "hypothesis": "La puerta sigue abierta. He dejado tu abrijo sobre la silla. Dime cuándo estás listo.", "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/emo_Pain/cbd57885898dcb7319c3c990916fb69d927f1755efbf2d40c8d97480bcc093b5.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": "cbeb685dbdcd3f7d6115de66222e5caf7f7c1b4e9dce72486ff41ef640dc5959", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cd5a8afa38179fc07d0ba86390713b96eeb01b22167badbf92801ce2e670e8c2", "generation_seconds_batch": 15.201315181911923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06261950731277466, "target_z": -0.3378906229759897, "ce": 6.054384231567383, "genuineness": 4.306513786315918, "identity": 0.34649646282196045, "hypothesis": "Iemand heeft het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "wer": 0.047619047619047616, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/cbeb685dbdcd3f7d6115de66222e5caf7f7c1b4e9dce72486ff41ef640dc5959.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": "cc2d589219ebb1cad10fd49b0abe618b4897a6f13d3df118fe51364d4ab4ba22", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "84ddfd47b9c9a490f2827b93d2277263f83378912257db84de0e2aa508231742", "generation_seconds_batch": 15.234316315036267, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4522429704666138, "target_z": 2.078124854271259, "ce": 5.7406487464904785, "genuineness": 4.293961524963379, "identity": 0.5410758256912231, "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": 0.03999999999999915, "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/emo_Pain/cc2d589219ebb1cad10fd49b0abe618b4897a6f13d3df118fe51364d4ab4ba22.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "cc76b0c0487ac2d4e381f554f2cb893c9dcfa5e75335df41854610e44628b3e5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9ecc63329c93c41b0beeb579076acb8001c18fbfb84034d49df57a8478e1d9cc", "generation_seconds_batch": 23.65698999899905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5378100872039795, "target_z": 0.48828123299831355, "ce": 6.197352409362793, "genuineness": 4.570107460021973, "identity": 0.43720924854278564, "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/emo_Pain/cc76b0c0487ac2d4e381f554f2cb893c9dcfa5e75335df41854610e44628b3e5.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": "cc8a9cb303dd3c32a41857f11c28b85cb73b1fb126c6d59cea23a243745c2232", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "573227c986d7060179b995ef54feb827bcbc81f6236582e3ade9ab0ee4cbd01f", "generation_seconds_batch": 15.450943839969113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5086021423339844, "target_z": 0.43749996113900236, "ce": 5.7684478759765625, "genuineness": 4.381826400756836, "identity": 0.7651089429855347, "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/emo_Pain/cc8a9cb303dd3c32a41857f11c28b85cb73b1fb126c6d59cea23a243745c2232.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "cca19c7f1b4a3e1945890d83af98a2d5e91e9474c2bb1f1f715977cd996a434a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b8566f79c5c311593fed2cf1d16eb14842add109f48534e7f2a26d87a7c3ea2e", "generation_seconds_batch": 22.60858390200883, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3827833831310272, "target_z": 0.21875000647683293, "ce": 6.0305562019348145, "genuineness": 4.758388519287109, "identity": 0.5513348579406738, "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/emo_Pain/cca19c7f1b4a3e1945890d83af98a2d5e91e9474c2bb1f1f715977cd996a434a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "cd0445a613903acfd20db2f253eef3a77b1a67f6b545b107a88b74fdb787bd19", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cd2d6f3066f33f85e220ebe1ba94bfd839e71d2cfefb1bd9a632d5de9733f365", "generation_seconds_batch": 22.36974727199413, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 6.316492557525635, "genuineness": 4.344170093536377, "identity": 0.5306068062782288, "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/emo_Pain/cd0445a613903acfd20db2f253eef3a77b1a67f6b545b107a88b74fdb787bd19.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "cd1e654069e373e478f95811b83fc24c95b58de26f37953ebe9bf068bf415f57", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7be5e88e61e0418a42a5c5ad5ed4c7e4c317a8bc26c7f3687ec2e438caa08ee3", "generation_seconds_batch": 26.16964961704798, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10530802607536316, "target_z": -0.2636718689279691, "ce": 5.824047088623047, "genuineness": 4.011539936065674, "identity": 0.4568253755569458, "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/emo_Pain/cd1e654069e373e478f95811b83fc24c95b58de26f37953ebe9bf068bf415f57.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "cd3dbd16f20a80e7656f691a45f6195f0fa7250af444b5082d91a395275e7c0d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e82ed86c2470d50665cc4c6c82d703569b8c34071346236bb0e1cc1086d6c5c9", "generation_seconds_batch": 21.364603562047705, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04352201521396637, "target_z": -0.3710937443327712, "ce": 6.316492557525635, "genuineness": 4.444586753845215, "identity": 0.5206804871559143, "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/emo_Pain/cd3dbd16f20a80e7656f691a45f6195f0fa7250af444b5082d91a395275e7c0d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "cd653ff453fffe86af82d109469d43a776d3b89ee4cca5106e2f22fe3d744d29", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "97002fbd5a33f497335f66ea5ed75c0eb811637a6de7d609f98e7e6fe7de8e90", "generation_seconds_batch": 15.275349429924972, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.748815894126892, "target_z": 2.593749980569501, "ce": 5.609594821929932, "genuineness": 4.068024158477783, "identity": 0.27505695819854736, "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/emo_Pain/cd653ff453fffe86af82d109469d43a776d3b89ee4cca5106e2f22fe3d744d29.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "cd73853114774eb0429e27933f0cc3f5c2001c40563ad07d7d74702df5d17ab9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "fd3a36ee3b7ba7023bf3ba38062efa5bbed2ada55ae0b36c6da2db153bcc7a95", "generation_seconds_batch": 21.689895846066065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.061306118965149, "target_z": 1.3984375080960412, "ce": 5.585766792297363, "genuineness": 4.287685871124268, "identity": 0.44573235511779785, "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/emo_Pain/cd73853114774eb0429e27933f0cc3f5c2001c40563ad07d7d74702df5d17ab9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "cd76bd30f946f2a3598787eda4054e336a24fcdbd5f454e50731fbdabb8de629", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0914014f671740d8edcb3d3f3e6e7243477e2a82a4351fd226c950e669c56220", "generation_seconds_batch": 22.520148242940195, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09407420456409454, "target_z": -0.2832031213567815, "ce": 6.229123115539551, "genuineness": 4.607763767242432, "identity": 0.7412887811660767, "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/emo_Pain/cd76bd30f946f2a3598787eda4054e336a24fcdbd5f454e50731fbdabb8de629.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "cdaaf42109ffd90afde8b646ee3987cfe9fa18b97920b6e9f1b1883ad9f006df", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "2cc0b3162676e5514f40e04ad1a10ea3ed76c441404797c1d0616a9528ab70c6", "generation_seconds_batch": 23.624915573047474, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.32071653008461, "target_z": 0.11083985559046021, "ce": 5.609594821929932, "genuineness": 4.042920112609863, "identity": 0.11343881487846375, "hypothesis": "Ik heb het kleine doosje gevonden. Het stond naast het oude raam. We kunnen het 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/emo_Pain/cdaaf42109ffd90afde8b646ee3987cfe9fa18b97920b6e9f1b1883ad9f006df.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "cdb7d99b1310fd2a8041da26d66d2d9cad747842445e2d989c5cd4f3837d9a4e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ab1d5c4b99ffc93a0d94109668eb3f70c02b2aca5a211db4e89e9666d4e3acf9", "generation_seconds_batch": 16.93266217398923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08845728635787964, "target_z": -0.29296876052485354, "ce": 6.1020402908325195, "genuineness": 4.770940780639648, "identity": 0.7499197125434875, "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/emo_Pain/cdb7d99b1310fd2a8041da26d66d2d9cad747842445e2d989c5cd4f3837d9a4e.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": "cde2322509a639bd060481a0364ce77b133c2ef7734a205849daf0d4b1b96099", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Escribí el número en el sobre.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Léelo bien... y comprueba las dos últimas cifras.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "565c8e8be09502acf94162faac47d3f8b99a390e75e374012ca9127658442d69", "generation_seconds_batch": 15.773294360027649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7085641622543335, "target_z": 0.7851562491903958, "ce": 5.935244083404541, "genuineness": 4.632867813110352, "identity": 0.6227093935012817, "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/emo_Pain/cde2322509a639bd060481a0364ce77b133c2ef7734a205849daf0d4b1b96099.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": "ce3236b6bfafd36d578f2ca59f0a4455c6c304cc14c1edabe8b008e8e417cdaf", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Iemand heeft het kleine blauwe stoeltje verplaatst.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond hier... naast de hoge boekenkast.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a29b880fd5b4d1e5eeaba09f5fc2b0b87319eb5a93cecfeb772be8b5d7bf5ad6", "generation_seconds_batch": 15.201315181911923, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16597065329551697, "target_z": -0.15820312135678147, "ce": 6.054384231567383, "genuineness": 4.068024158477783, "identity": 0.43599778413772583, "hypothesis": "Iemand heeft het het kleine blauwe stoeltje verplaatst. Het stond hier naast de hoge boekenkast. We zetten het naar de lunch terug.", "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/emo_Pain/ce3236b6bfafd36d578f2ca59f0a4455c6c304cc14c1edabe8b008e8e417cdaf.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "ce8a42a5375d6fec070b97df13324efdf418509c56b3b8a645df138a59b9f58d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cfc14f173f81cea3954ec174d23091d232307f9e772c7de513c537fa75d3071c", "generation_seconds_batch": 21.9986300150631, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.030041426420211792, "target_z": -0.3945312524288124, "ce": 5.911416053771973, "genuineness": 4.607763767242432, "identity": 0.43475231528282166, "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/emo_Pain/ce8a42a5375d6fec070b97df13324efdf418509c56b3b8a645df138a59b9f58d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "cf0c6b7338031906ba1df6838fef1d5ed6457a8697c0a229c3aa5082bfbcb76f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f39bedad72abaefa5e1375031b86481fbed0c1f9ff61a13fd9d378b3029f0532", "generation_seconds_batch": 23.144475175999105, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18057462573051453, "target_z": -0.1328124854271259, "ce": 5.871703147888184, "genuineness": 4.532451152801514, "identity": 0.7463123202323914, "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/emo_Pain/cf0c6b7338031906ba1df6838fef1d5ed6457a8697c0a229c3aa5082bfbcb76f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "cf4093cf0daccee1f4f025e3a09bc3689c5c003f0509d1c6dce3630b13d257bc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "18d13de2ff8267d1ab3af9e499f4df1cd124d2f67420c4c2529179507cce6699", "generation_seconds_batch": 24.086277142050676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2050989866256714, "target_z": 1.6484374562813777, "ce": 5.816103935241699, "genuineness": 4.507347106933594, "identity": 0.6142484545707703, "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/emo_Pain/cf4093cf0daccee1f4f025e3a09bc3689c5c003f0509d1c6dce3630b13d257bc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "cf58d72065ff7ab5fb9362b7867eb1fc6ffee147008d8219288efb2f381e8be2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "189305227ff8015f0e31d33ea94d597ecdc2f32f3de64323663110ae93899374", "generation_seconds_batch": 25.119574659969658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12440551817417145, "target_z": -0.23046874757118765, "ce": 6.229123115539551, "genuineness": 4.65797233581543, "identity": 0.26504093408584595, "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/emo_Pain/cf58d72065ff7ab5fb9362b7867eb1fc6ffee147008d8219288efb2f381e8be2.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "0354f38bb34346509826561e8f6acb86bffa5fb0e3454ede1726622ec5620bbf", "generation_seconds_batch": 17.549043214996345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.037905097007751465, "target_z": -0.38085938350084325, "ce": 6.316492557525635, "genuineness": 4.381826400756836, "identity": 0.6087255477905273, "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/emo_Pain/cf68e6506caaaad7e7371e72de9d5417fef89211300487ff83acd2484eaab510.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "cfab3062e241e265efdfde8fef4e1cbda39e321b3b92f1eb08386a36e89ef670", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e7f6c652da149d3c0a0fcd15881181db94d0ceb15dd926942c9972343f189c30", "generation_seconds_batch": 22.52575620589778, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6771094799041748, "target_z": 0.7304687734785194, "ce": 5.796247482299805, "genuineness": 4.419482707977295, "identity": 0.15177279710769653, "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/emo_Pain/cfab3062e241e265efdfde8fef4e1cbda39e321b3b92f1eb08386a36e89ef670.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": "cfbe4d9b19410d0eade5b20cb5b3b731096f5fb245e4f864a931bd0ea6a48e61", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Natural clear speech.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Natural clear speech.) [3.2 seconds duration] Appelle-moi quand tu auras un moment.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "6a9a3bfec32358b7589a56b0911a2c5f48616ebd8f5e1e97dca97e8125ec837d", "generation_seconds_batch": 17.29066447902005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15923035144805908, "target_z": -0.16992188835846794, "ce": 5.577824115753174, "genuineness": 4.557555198669434, "identity": 0.7987778186798096, "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/emo_Pain/cfbe4d9b19410d0eade5b20cb5b3b731096f5fb245e4f864a931bd0ea6a48e61.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": "d0bf6e75a409c3284643c4c02d7c7f3e146ea162a76816be0459c9083f295a30", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c9fc450f31019ae0d28eaf8574c722d0b0cdc1f49b0a29805c77e9cbb8b97440", "generation_seconds_batch": 23.057104796054773, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2552795112133026, "target_z": -0.002929704704087493, "ce": 6.046441555023193, "genuineness": 4.582659721374512, "identity": 0.6799138188362122, "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/emo_Pain/d0bf6e75a409c3284643c4c02d7c7f3e146ea162a76816be0459c9083f295a30.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "b1028ded9abde8cb9a514e525d1c036567ebf18918d1cd6cf5cb3a47bc9b7a7a", "generation_seconds_batch": 17.480708557064645, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04913891851902008, "target_z": -0.3613281310720309, "ce": 6.403861999511719, "genuineness": 4.20609712600708, "identity": 0.14114564657211304, "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": 0.03999999999999915, "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/emo_Pain/d11caac31fa3732e2517803823f1b039213fb31e9438619fd83d8b232c894b7d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "d16045b4e8ec1bc9f753e30c6ce49e753ac749b488652c47ee585c77c909e925", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d8a410894bc774b8da50581bc8c13560b262b343e6db19a05598582c543cf01a", "generation_seconds_batch": 18.37425973196514, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7804605960845947, "target_z": 0.9101562232830641, "ce": 6.0146708488464355, "genuineness": 4.7081804275512695, "identity": 0.814087986946106, "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/emo_Pain/d16045b4e8ec1bc9f753e30c6ce49e753ac749b488652c47ee585c77c909e925.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "d183ebeec3b9c044c273bdb9f43424ad493ca5e0c43675aa4c937ea4500be7ce", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "01d10f51fe42ee693cab49e11e56e823dfdba1a0e4b96117e4fc82f8666b22ac", "generation_seconds_batch": 24.521255424944684, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.213237762451172, "genuineness": 4.20609712600708, "identity": 0.4682125151157379, "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/emo_Pain/d183ebeec3b9c044c273bdb9f43424ad493ca5e0c43675aa4c937ea4500be7ce.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "d22b50684140f113a2652fe4c480c5f5036731762297560fa24b599e57a2faff", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4a9707db341ed0572fc491523bd59c7b443e283c54ec41b9688033cc3c477428", "generation_seconds_batch": 22.821842935984023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38615351915359497, "target_z": 0.2246093640703444, "ce": 5.871703147888184, "genuineness": 4.570107460021973, "identity": 0.2773062586784363, "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/emo_Pain/d22b50684140f113a2652fe4c480c5f5036731762297560fa24b599e57a2faff.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "5dd35194b3028f1dec998f7f215551c4f9346fd8edcf1a674b4e3f023ef73349", "generation_seconds_batch": 16.675313247949816, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.058125972747802734, "target_z": -0.34570313431044736, "ce": 5.990842819213867, "genuineness": 4.344170093536377, "identity": 0.7316434383392334, "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/emo_Pain/d2319f0ba4d4a13f099c61fd2524d2cd47bb09bca092815ed8a29e7f5b0c527c.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": "d27c428eb34bfec2fcbda9e531a02769cd18f458880f7d1768831e8e9017e780", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0711cac602d6ddb190bb72a2b079ab0f2583c59ce348b410f896b36bf005aa37", "generation_seconds_batch": 23.107928665005602, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16597065329551697, "target_z": -0.15820312135678147, "ce": 6.197352409362793, "genuineness": 4.356721878051758, "identity": 0.7167658805847168, "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/emo_Pain/d27c428eb34bfec2fcbda9e531a02769cd18f458880f7d1768831e8e9017e780.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "d28d5909d3309d07151cb41709448e9fc784dfe811405a7c249e87016a5aac17", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "06e57a7b5ca52c42da1e583cfb56cac1bb5de076929f6f4bc90b74d8d0b5c59c", "generation_seconds_batch": 17.573600075091235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 5.772419452667236, "genuineness": 4.532451152801514, "identity": 0.6269121170043945, "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/emo_Pain/d28d5909d3309d07151cb41709448e9fc784dfe811405a7c249e87016a5aac17.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": "d2b7514a4baf72ceaed5b8568ea6e533fa52d27c78ee0f553d1ba1e013c58e8c", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik schreef het nummer op de envelop.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lees het goed... en controleer de laatste twee cijfers.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "43c3000ad66ebe986c264f463c0d8cf7abb8af4fab5c65a48b6d1fc37b2fbec9", "generation_seconds_batch": 22.78990407299716, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23927132785320282, "target_z": -0.030761718041596397, "ce": 5.911416053771973, "genuineness": 4.394378662109375, "identity": 0.19094230234622955, "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/emo_Pain/d2b7514a4baf72ceaed5b8568ea6e533fa52d27c78ee0f553d1ba1e013c58e8c.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": "d2be902af6e36a0a444f39b485b541089e77ad126094a38dca2c8916127e59e3", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Natural clear speech.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Natural clear speech.) [3.2 seconds duration] Nous la remettrons après le déjeuner.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "2928fbd84041ab9fe43834f1fc119eafcbcc6f8fd49327c8f1bf1196363f0e29", "generation_seconds_batch": 17.164312410983257, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06823641061782837, "target_z": -0.3281250097152494, "ce": 6.165581703186035, "genuineness": 4.469690799713135, "identity": 0.65479975938797, "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/emo_Pain/d2be902af6e36a0a444f39b485b541089e77ad126094a38dca2c8916127e59e3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "d2c92705d349ff67006def21217147c118f7168e60bdf5dd48605faaadbe4730", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a9f6ed564f91e2cb4c15f5bf4218c48964d76b76fb6c18029313df1792189d91", "generation_seconds_batch": 23.208367681945674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10081449151039124, "target_z": -0.27148438026242677, "ce": 6.23706579208374, "genuineness": 4.243753433227539, "identity": 0.6228122115135193, "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/emo_Pain/d2c92705d349ff67006def21217147c118f7168e60bdf5dd48605faaadbe4730.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "d311ccf4e38de47c398c7830fe32699e2954593fa31973870052118d3918a19c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6c47558fec53f47b401ec3ce088838cf7b55d3b2730efa77235dd7e717d9accb", "generation_seconds_batch": 22.08014843193814, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6052130460739136, "target_z": 0.6054687993858512, "ce": 5.657250881195068, "genuineness": 4.68307638168335, "identity": 0.76474928855896, "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/emo_Pain/d311ccf4e38de47c398c7830fe32699e2954593fa31973870052118d3918a19c.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": "d3138e376837355828021feda41f8f24b2013aac5e4fef993712f46e2bfc64dd", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "008101e0f114443ca332cb5a98a04dd693b9f8ca8c15c50e8d40823bee2644bd", "generation_seconds_batch": 21.642902194987983, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3844684362411499, "target_z": 0.22167965936625691, "ce": 5.5539960861206055, "genuineness": 4.168440818786621, "identity": 0.6214423179626465, "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/emo_Pain/d3138e376837355828021feda41f8f24b2013aac5e4fef993712f46e2bfc64dd.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "d366fee16ad9e63d252ed8b4ec9202e90dbab7973395ecb0369300e9ed10b18c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b35016527dc91b969f265d7d1944ea81c6f94ccba06ba6aac52a363601565c3c", "generation_seconds_batch": 21.29163441399578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.16260050237178802, "target_z": -0.1640625048576247, "ce": 5.831989288330078, "genuineness": 4.582659721374512, "identity": 0.7245673537254333, "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/emo_Pain/d366fee16ad9e63d252ed8b4ec9202e90dbab7973395ecb0369300e9ed10b18c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "d3ee6f60d29bc4b244fdf55b38244a0227d9568127c12f0fea8abe3df995df45", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6b8ca6c1fece6634f1eb0a1a0527d826f1a52465b9cedddb48ed95b8fcc2ebda", "generation_seconds_batch": 22.53379259898793, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5254528522491455, "target_z": 0.4667968009212233, "ce": 5.585766792297363, "genuineness": 4.545003414154053, "identity": 0.5841730833053589, "hypothesis": "I found the little books. It was b beside 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/emo_Pain/d3ee6f60d29bc4b244fdf55b38244a0227d9568127c12f0fea8abe3df995df45.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "59c62780f530786b275e249078d4d8d89e09457793941399b2b6bcbd5c47a6ea", "generation_seconds_batch": 16.467003004974686, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.036781713366508484, "target_z": -0.38281251133445765, "ce": 6.022613525390625, "genuineness": 4.557555198669434, "identity": 0.7397247552871704, "hypothesis": "J'ai écrit le numéro sur l'un l'enveloppe, lis 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/emo_Pain/d3fab0ead1fc9dcf899f59bddb0c9d4b54373ba1829b6dc49f77fc4027ce1740.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "d41eda1174ddce8e5ebbe3a9f26dd5f40bed13ee611241690163f8ff451f1006", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "40069f9f9e7b3deb5e38acf10381a4bd124a888444d68d1a3b5276e7a3269b15", "generation_seconds_batch": 24.39616763300728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 6.348263263702393, "genuineness": 4.457139015197754, "identity": 0.38995420932769775, "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/emo_Pain/d41eda1174ddce8e5ebbe3a9f26dd5f40bed13ee611241690163f8ff451f1006.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "d42cd2a14f56b1edfa564aac54fef5de0c8603167d587dd566c7b7fad21959e2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 129, "seconds_generated": 10.32, "mp3_sha256": "ad5b6023f2275933c1512e15c384911d9f48a01ed47b4ceafe288d687865aef8", "generation_seconds_batch": 22.27053001697641, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8566606044769287, "target_z": 2.7812500453378304, "ce": 5.39911413192749, "genuineness": 4.105680465698242, "identity": 0.1078043282032013, "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": 0.120000000000001, "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/emo_Pain/d42cd2a14f56b1edfa564aac54fef5de0c8603167d587dd566c7b7fad21959e2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "d4a8f60d680f684a65375117c06edcca969ccc57d8f3f6180f6ea8daaec14d4c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d4ff8fbdd147b04d337e9a1c0f90aa57b5710752c4ee30b003f15507f4547556", "generation_seconds_batch": 17.37159056600649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22789707779884338, "target_z": -0.050537121164859956, "ce": 5.919358730316162, "genuineness": 4.7332844734191895, "identity": 0.8217439651489258, "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/emo_Pain/d4a8f60d680f684a65375117c06edcca969ccc57d8f3f6180f6ea8daaec14d4c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "d4c2cfe4fe84d5389b7e07d12434165fa90dbb2b8ea9482d54a235645deb62a2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9175e8818824699e29843b40dd087acb9089a277ce31756a61fa598e72186928", "generation_seconds_batch": 22.540301393019035, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08059361577033997, "target_z": -0.30664062945282267, "ce": 5.895530700683594, "genuineness": 4.482243061065674, "identity": 0.7356104850769043, "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/emo_Pain/d4c2cfe4fe84d5389b7e07d12434165fa90dbb2b8ea9482d54a235645deb62a2.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": "d4c62459d613978bf6cb9b403f5e703ad2d8ed65fd95180a2d556b4705dad2f6", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "cfb5db14dfa60f267a5a7fcfb7efcb47591819070b27d5ef2a9906bcb48ece95", "generation_seconds_batch": 22.05796927504707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2140860557556152, "target_z": 1.664062478950293, "ce": 5.299830436706543, "genuineness": 4.3504462242126465, "identity": 0.6445000171661377, "hypothesis": "Deine Nachricht kam, als sie schon weg war. Ich lass sie zweimal. Dann legte ich das Handy weg. Sag 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/emo_Pain/d4c62459d613978bf6cb9b403f5e703ad2d8ed65fd95180a2d556b4705dad2f6.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": "d51d5ecf024f862ac00f15e25392ee87c2d4707723fe8aaa0caa8659a5579056", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Alguien ha movido la pequeña silla azul.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba aquí... junto a la estantería alta.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9619c6b008d12cd2cfe12c2b003e5f647708237ea614d48aad7153e5f9423e7d", "generation_seconds_batch": 16.253656171029434, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41873160004615784, "target_z": 0.28124999352316704, "ce": 5.8875885009765625, "genuineness": 4.519899368286133, "identity": 0.5764698386192322, "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/emo_Pain/d51d5ecf024f862ac00f15e25392ee87c2d4707723fe8aaa0caa8659a5579056.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "d5452c51f8cabd70942696b4f4d84ab37153d6cda5f12c4da199edefdd9b88d2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "74e93d98ca568667b72a6ce4952d14d0a39f18e769a16006d2ce548194c2eec4", "generation_seconds_batch": 26.16772031004075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6456547975540161, "target_z": 0.675781297766643, "ce": 5.605623245239258, "genuineness": 4.406930446624756, "identity": 0.5854076147079468, "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/emo_Pain/d5452c51f8cabd70942696b4f4d84ab37153d6cda5f12c4da199edefdd9b88d2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "d553b56a9c62f5855bfa2d61931788cf2634ed1b8b89638abfc7cc5bb13c9f8e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "7824865cacda19fb77e2b805afb3e3765615a6cca5a0b5e346dc05935564b2aa", "generation_seconds_batch": 22.934741495992057, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12552890181541443, "target_z": -0.22851561973757323, "ce": 5.935244083404541, "genuineness": 4.632867813110352, "identity": 0.9010592103004456, "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/emo_Pain/d553b56a9c62f5855bfa2d61931788cf2634ed1b8b89638abfc7cc5bb13c9f8e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "d55cd108a70e74bce500cb64a7562fbf786515278862e5964eb0f43678d1bb1f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f3d0e2d1b709689934d7527ddbc6daed3540d226789d2d7ddf1ee7634fc445f5", "generation_seconds_batch": 24.89237240003422, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18113631010055542, "target_z": -0.13183593446398456, "ce": 5.998785495758057, "genuineness": 4.758388519287109, "identity": 0.6413593292236328, "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/emo_Pain/d55cd108a70e74bce500cb64a7562fbf786515278862e5964eb0f43678d1bb1f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "d569b7adc7d1fd97cfe31c48a4166ac5debad916fd0d27b3c5713614362ee663", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6ebdede2b7ef2592639f57f12de6e7c4061844d307ed8a7661cbd7e30007df80", "generation_seconds_batch": 22.002731401007622, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14911991357803345, "target_z": -0.1875000129536659, "ce": 6.1020402908325195, "genuineness": 4.15588903427124, "identity": 0.5865350961685181, "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/emo_Pain/d569b7adc7d1fd97cfe31c48a4166ac5debad916fd0d27b3c5713614362ee663.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": "d58ec1687f36a1fc0bd057ce96fedbfbd48c03db49b7d84bb9949e286db501ab", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "d9880cbc95fc50328b88e437c512ff11f017c85e26828c5324962b8fed98fdb9", "generation_seconds_batch": 22.614626239053905, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.4881912469863892, "target_z": 2.14062494494692, "ce": 5.502368927001953, "genuineness": 4.218649387359619, "identity": 0.5773122906684875, "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": 0.03999999999999915, "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/emo_Pain/d58ec1687f36a1fc0bd057ce96fedbfbd48c03db49b7d84bb9949e286db501ab.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "d5bf78eb54398c5ac35e89ae54c1022130713452a5e2b85252a28d501baedf8b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d5f47ba4d1f127a8b539c6b7acc9bb669219bc8121d4c95299959b29b8e7881e", "generation_seconds_batch": 23.225750015000813, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2002338021993637, "target_z": -0.0986328131072031, "ce": 5.689021587371826, "genuineness": 4.432034492492676, "identity": 0.631924033164978, "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/emo_Pain/d5bf78eb54398c5ac35e89ae54c1022130713452a5e2b85252a28d501baedf8b.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": "d5da7d58844a9705dc9c976c5ad08ecb9646d07f6796d2978640c10c9f1dd1c3", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "557bb58d011c7cdad5c9d3f6eb2ee3c9c53886324ce3290bd675924b0a85091e", "generation_seconds_batch": 19.215147578972392, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1511766910552979, "target_z": 1.55468752752654, "ce": 6.054384231567383, "genuineness": 4.306513786315918, "identity": 0.5048012733459473, "hypothesis": "Your message arrived after I had left. I read it twice before putting the phone away. Let me know when we can talk.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/d5da7d58844a9705dc9c976c5ad08ecb9646d07f6796d2978640c10c9f1dd1c3.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": "d5fd2b72f42dc8b03bf3252f3b0ffd7cc3b34519b3ee7d14a2442c034788eb9a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "532d85c7c2a7ed43f2998b18f349516e8a76a1f4282b818afafa6fe3ab3bce62", "generation_seconds_batch": 26.607766156084836, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.30189988017082214, "target_z": 0.07812500971524941, "ce": 5.935244083404541, "genuineness": 3.9927117824554443, "identity": 0.5916390419006348, "hypothesis": "The keys are on the kitchen table. I checked the drawer and looked beside 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/emo_Pain/d5fd2b72f42dc8b03bf3252f3b0ffd7cc3b34519b3ee7d14a2442c034788eb9a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "d642abe8ce38fef4d67c199b640db7155f847e6dee1be046145b9947c45ea284", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "d4f761dd49ad20f4c3e3de155423e73ab7e32b24f18d60bbab7897728242b548", "generation_seconds_batch": 24.032599430996925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8860585689544678, "target_z": 1.0937500841988281, "ce": 5.530168056488037, "genuineness": 4.243753433227539, "identity": 0.5405627489089966, "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/emo_Pain/d642abe8ce38fef4d67c199b640db7155f847e6dee1be046145b9947c45ea284.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "d65361b3528e5dfc87c1dbc0c6787a3c634230e19defae0a5ef1a056e3571fcb", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9278f926c81f22b2bb3b44eff5f8ee08f362d6ad1bc4969a9c80da2dbb8219db", "generation_seconds_batch": 17.58817168790847, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 5.903473377227783, "genuineness": 4.325342178344727, "identity": 0.4243316948413849, "hypothesis": "Ik heb een kleine doosje gevonden. Het stond naast het oude raam. We kunnen ik er 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/emo_Pain/d65361b3528e5dfc87c1dbc0c6787a3c634230e19defae0a5ef1a056e3571fcb.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "d7ebe98295e665cbbb52d6af61ccb9dd47e35cfd450f81440fafed52b516f9d5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "608d44b515b48b98ec624f3d3d173ad7d8c61fe00d58ad0d2812d12c4aa13fd0", "generation_seconds_batch": 20.550634487066418, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14743484556674957, "target_z": -0.19042969175042163, "ce": 6.070269584655762, "genuineness": 4.796044826507568, "identity": 0.6571252942085266, "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/emo_Pain/d7ebe98295e665cbbb52d6af61ccb9dd47e35cfd450f81440fafed52b516f9d5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "d7f5668fe2d6d708f1a33243d82fac918152ee0294b1a7d08268d0a9e627abd9", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6e0700758d3ceff3b32261bc406ff8e3ac60e2d101b0f6c1341d068f320202d3", "generation_seconds_batch": 17.37159056600649, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06261950731277466, "target_z": -0.3378906229759897, "ce": 6.078212261199951, "genuineness": 4.607763767242432, "identity": 0.7136726379394531, "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/emo_Pain/d7f5668fe2d6d708f1a33243d82fac918152ee0294b1a7d08268d0a9e627abd9.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "d80fdc539c3b1b3aad060a2be3492820c9de939a7772a1471d5c63abb15bbc3b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a84df6c99d2181aa93332409bacd36d56a8ce62a03c32a143b4c1cc972ff29df", "generation_seconds_batch": 30.399186848080717, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09969110786914825, "target_z": -0.27343750809604117, "ce": 6.2767791748046875, "genuineness": 4.7834930419921875, "identity": 0.5421118140220642, "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/emo_Pain/d80fdc539c3b1b3aad060a2be3492820c9de939a7772a1471d5c63abb15bbc3b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "d83f349ae6e2520f87c048cf45554dfea8b905776c949b043bd764e419bb0c00", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "8e6764cf28fc5199e50df30e74f013e2b778e39aaf09ca461674f5b5f42310d2", "generation_seconds_batch": 22.790873651974835, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4749007225036621, "target_z": 0.3789062815745606, "ce": 5.847874641418457, "genuineness": 4.7332844734191895, "identity": 0.6844195127487183, "hypothesis": "La porte est encore ouverte. J'ai poussé 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/manual-comparison-v1/audio/emo_Pain/d83f349ae6e2520f87c048cf45554dfea8b905776c949b043bd764e419bb0c00.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "d84087759a4d3e14c6525ed72c8b2ea9f94a94b01ba6fa392324d17a8561fa1f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d15ee3f39b742dfdb22062a820de0e234030e1e61849055aed73444e26f9c1a2", "generation_seconds_batch": 25.888174471096136, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3732346296310425, "target_z": 0.20214843284477632, "ce": 6.23706579208374, "genuineness": 4.168440818786621, "identity": 0.740969181060791, "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/emo_Pain/d84087759a4d3e14c6525ed72c8b2ea9f94a94b01ba6fa392324d17a8561fa1f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "d8c3b3c081e907e2a2275b462ea71d61427d929c74531c989210b95ae8df3ec4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8a86c3342b047cdfcb63b9dae35b11b38974b3ff69232fd26bd8d523b60898f9", "generation_seconds_batch": 17.852927301893942, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.41086792945861816, "target_z": 0.2675781245951979, "ce": 6.292664527893066, "genuineness": 4.306513786315918, "identity": 0.5375782251358032, "hypothesis": "Encontrado la cajita. Estaba junto a la ventana. Vieja. Podemos mirar juntos lo que hay dentro.", "wer": 0.058823529411764705, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/d8c3b3c081e907e2a2275b462ea71d61427d929c74531c989210b95ae8df3ec4.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": "d8d7e0c7016f1017b2d4e78830b17c6c5feea8739e2c426d395dcb185c6b716a", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De weg gaat linksaf na de brug.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Loop verder... tot aan het kleine houten hek.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4c827644c5f6f67214fc45a48e3d3818bb6d105e858b3ee885a5490338e357d0", "generation_seconds_batch": 22.488499677041546, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0972543954849243, "target_z": 1.4609375987717024, "ce": 5.863759994506836, "genuineness": 4.2249250411987305, "identity": 0.3298757076263428, "hypothesis": "De weg gaat gaat linksaf naar de brug. Loop 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/emo_Pain/d8d7e0c7016f1017b2d4e78830b17c6c5feea8739e2c426d395dcb185c6b716a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "d942b8fc81e45af53315854d79115a3c3088249547d2720d72e6f2e708021ff0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "9c3138d5a7e84f19d5e51d077d575b00afaded0b337edc64aa63689e5a89ab96", "generation_seconds_batch": 22.942274622037075, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.8027381896972656, "target_z": 2.687499909324339, "ce": 5.1767191886901855, "genuineness": 3.798154592514038, "identity": 0.5108977556228638, "hypothesis": "J'ai trouvé la petite boîte. Elle était putain de boîte. Elle était près de la vieille fenêtre. Nous pouvons regarder ensemble ce qu'elle contient.", "wer": 0.2631578947368421, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/d942b8fc81e45af53315854d79115a3c3088249547d2720d72e6f2e708021ff0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "d993a7526c3694d2558b0677773892ee435c1f3d02b36f2e8eeae91e7bd13b61", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e9efb07efbb58cc688ede7e9755dbdd8f53991409fec7ac8737a7740e8d9ffbf", "generation_seconds_batch": 16.084019648027606, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9669420719146729, "target_z": 1.2343750809604117, "ce": 5.780362129211426, "genuineness": 4.268857479095459, "identity": 0.7318317890167236, "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/emo_Pain/d993a7526c3694d2558b0677773892ee435c1f3d02b36f2e8eeae91e7bd13b61.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": "d9b399bea8a1b3a1a9f42f12a68e613d4c5e394a19fe0bdac34f7d766cc2b98e", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Las llaves están en la mesa de la cocina.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Miré en el cajón... y luego junto a la ventana.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "bce461840a02352e9f869c9ae1cc135f54f1044597b6f09bf1a208186d859bdb", "generation_seconds_batch": 15.450943839969113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8590973615646362, "target_z": 1.0468750161920823, "ce": 5.831989288330078, "genuineness": 4.262581825256348, "identity": 0.6670866012573242, "hypothesis": "Más llaves. Están en la mesa de la cocina. Miré 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/emo_Pain/d9b399bea8a1b3a1a9f42f12a68e613d4c5e394a19fe0bdac34f7d766cc2b98e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "d9dda4d68bfa233d3a4e0cebac0b19fc90f9dba3d0dddda6edff8f3fea857801", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4077935de4178368349c2d1be99a13c8a191347adf948ef3fb255c75de93040a", "generation_seconds_batch": 18.31059727095999, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.18506814539432526, "target_z": -0.125, "ce": 6.1020402908325195, "genuineness": 4.720732688903809, "identity": 0.7109794020652771, "hypothesis": "J'ai trouvé la petite boîte. Elle était près de 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/emo_Pain/d9dda4d68bfa233d3a4e0cebac0b19fc90f9dba3d0dddda6edff8f3fea857801.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "d9f7f625d1b4c8cb16b1472a60eb3a0c66b6e47553809f1bf58c0d993249a848", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "929b4df360ceb5babd23217ac1d6db61fe6e9353f2e52d75aec09b1d33a45baf", "generation_seconds_batch": 21.53002063499298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07385332882404327, "target_z": -0.31835937054717733, "ce": 6.157639026641846, "genuineness": 4.65797233581543, "identity": 0.7912133932113647, "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/emo_Pain/d9f7f625d1b4c8cb16b1472a60eb3a0c66b6e47553809f1bf58c0d993249a848.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "da34e933022bade540aa2a6fd2154d186a210d76266cb9ccbfbe1b9cc45dab57", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7319fe00bec304736dc816823f32e59b9892d68486b132e03c9c2dc7b2132d10", "generation_seconds_batch": 24.3197509549791, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.17158755660057068, "target_z": -0.14843750809604117, "ce": 6.070269584655762, "genuineness": 4.457139015197754, "identity": 0.10779894888401031, "hypothesis": "Ik heb het klein 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/emo_Pain/da34e933022bade540aa2a6fd2154d186a210d76266cb9ccbfbe1b9cc45dab57.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "da6646e402cd55a30e53f97ed1a3cf42d9e50fe76aa8f8bb3338c716efd4ef4d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4cba028e82d0e3fd8a80d0f190ca6b6fed967a168f678783079b4bc74082c0bb", "generation_seconds_batch": 21.330394521937706, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2780280113220215, "target_z": 0.03662110154243963, "ce": 6.125868320465088, "genuineness": 4.250029563903809, "identity": 0.6002867817878723, "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/emo_Pain/da6646e402cd55a30e53f97ed1a3cf42d9e50fe76aa8f8bb3338c716efd4ef4d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "daa8b3ce274bdbb2add1b296ac12848ad1effb7e194b8d320d0dc3919046b361", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "85e85cb4820db91cdaab1d95a9e230ec717446a7a439f4b0cc95f49fb75d518c", "generation_seconds_batch": 24.39616763300728, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1154184639453888, "target_z": -0.24609374433277117, "ce": 6.41180419921875, "genuineness": 4.720732688903809, "identity": 0.5382123589515686, "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/emo_Pain/daa8b3ce274bdbb2add1b296ac12848ad1effb7e194b8d320d0dc3919046b361.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "db131c0ceb425bce527771c82598b1cc296bc4469e7ab7290904fe032167b3db", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6a2db11fb09f3a9078f6a6e65519f4bebeb28f6f9073498f4b4f262fb7b80950", "generation_seconds_batch": 22.32786624704022, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07947023212909698, "target_z": -0.30859375728643706, "ce": 6.070269584655762, "genuineness": 4.6203155517578125, "identity": 0.5468653440475464, "hypothesis": "He encontrado la cajita. Estaba junto a la 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/emo_Pain/db131c0ceb425bce527771c82598b1cc296bc4469e7ab7290904fe032167b3db.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "db17bbadf03ba32dd7e116ae9003e48bd1aa91a887af4673f90472151ab1ca3b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We kunnen er samen in kijken.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Natural clear speech.) [3.8 seconds duration] Het stond naast het oude raam.\n(Natural clear speech.) [3.2 seconds duration] We kunnen er samen in kijken.", "seed": 1702, "scale": 1.0, "cfg": 5.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "78a53c1cf002c36b007895cf5902baa82842e0c7b0a64809bfe8b0c9135a9235", "generation_seconds_batch": 23.031720163999125, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.5331264734268188, "target_z": 2.2187498510328423, "ce": 5.903473377227783, "genuineness": 4.3127899169921875, "identity": 0.21230243146419525, "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/emo_Pain/db17bbadf03ba32dd7e116ae9003e48bd1aa91a887af4673f90472151ab1ca3b.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "072c60e2ccc9a3512be14e18bbc9a2b961267413dc52bdefdd525727c51973d6", "generation_seconds_batch": 16.90793301607482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 6.38797664642334, "genuineness": 4.20609712600708, "identity": 0.3725934624671936, "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/emo_Pain/db192910e9ae67de251fb2a990940eab881f66f627c78b50b638ce61f5a63679.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "db597916da0f5027f1daa09b41dcaefbd2e4d0c2ab34a32c18afe4949b7dbcc2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e34c04e5e7fdf875edf0ae8df275e3b2561bd9a5634389cb0f3b490063a22199", "generation_seconds_batch": 21.64186190092005, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 6.054384231567383, "genuineness": 4.846253395080566, "identity": 0.7684886455535889, "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/emo_Pain/db597916da0f5027f1daa09b41dcaefbd2e4d0c2ab34a32c18afe4949b7dbcc2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "db6fdc28834cf6c1d7f18c2cf2ae65f593c0ad318188b2fd334fb927a4a97788", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] De deur staat nog open.\n(Natural clear speech.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Natural clear speech.) [3.2 seconds duration] Zeg maar wanneer je klaar bent.", "seed": 1702, "scale": 0.5, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "95404ed225ce07ccdc8a2b301f1575713446f75ad1220d870701e0537516f5d1", "generation_seconds_batch": 24.174109110957943, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3277376592159271, "target_z": 0.12304687216638559, "ce": 5.792276382446289, "genuineness": 4.444586753845215, "identity": 0.4196731150150299, "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/emo_Pain/db6fdc28834cf6c1d7f18c2cf2ae65f593c0ad318188b2fd334fb927a4a97788.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "dbd44d99f14df2ec15ca61d4e2742565eddd3957030f7c84ac2bf52ef551c7e4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "45e0bd4b8b951c8e36843ae13ac495fcef7651bd01769c2644401ffcb5a562bf", "generation_seconds_batch": 22.469890577020124, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9489679336547852, "target_z": 1.203125035622581, "ce": 5.303802013397217, "genuineness": 4.570107460021973, "identity": 0.5377484560012817, "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/emo_Pain/dbd44d99f14df2ec15ca61d4e2742565eddd3957030f7c84ac2bf52ef551c7e4.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": "dd129f9b45595d17a9aa8579627c1040f1d68b9771070b16a1aa1fd00085e76b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Natural clear speech.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Natural clear speech.) [3.2 seconds duration] Je t’attendrai de l’autre côté.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5ae9865ce8816eb7142b37b53067b621c19220c6f3b96a1c7b0cacc1891a2afa", "generation_seconds_batch": 16.215985237038694, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8321361541748047, "target_z": 0.9999999481853364, "ce": 6.046441555023193, "genuineness": 4.482243061065674, "identity": 0.7227381467819214, "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/emo_Pain/dd129f9b45595d17a9aa8579627c1040f1d68b9771070b16a1aa1fd00085e76b.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": "dd3c17d10647871af6951b06e2b7c4eccf3e0aa798a495c9920336b28294e2fd", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Jemand hat den kleinen blauen Stuhl umgestellt.\n(Natural clear speech.) [3.8 seconds duration] Er stand hier... neben dem hohen Bücherregal.\n(Natural clear speech.) [3.2 seconds duration] Wir stellen ihn nach dem Essen zurück.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 129, "seconds_generated": 10.32, "mp3_sha256": "b02e5ae1e7b4238d11262c3142ba0e2ac69c7de4043b01347bab445fe13b376e", "generation_seconds_batch": 17.221479989006184, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.054384231567383, "genuineness": 4.582659721374512, "identity": 0.7376912832260132, "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/emo_Pain/dd3c17d10647871af6951b06e2b7c4eccf3e0aa798a495c9920336b28294e2fd.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": "dd5ea5fc36fc86ce6206cf3da7383d319ab13a0cba6def10751eb57336931eab", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Prends-les quand tu partiras.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Les clés sont sur la table de la cuisine.\n(Natural clear speech.) [3.8 seconds duration] J’ai regardé dans le tiroir... puis près de la fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Prends-les quand tu partiras.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "3c35434f69fe17f353416da17820a26377c241e29e4e27edfa141cfcd3ccd305", "generation_seconds_batch": 17.465707438997924, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.33279287815093994, "target_z": 0.13183593446398456, "ce": 5.839932441711426, "genuineness": 4.325342178344727, "identity": 0.6573281288146973, "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/emo_Pain/dd5ea5fc36fc86ce6206cf3da7383d319ab13a0cba6def10751eb57336931eab.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "dd98b0c40c19f6be6e9357277762a4e3af0fd2521079526eca09459a735552f7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f930f4cce14eaf2ecfd5557b788166564c2e642b391fc0171e670b61b9935d47", "generation_seconds_batch": 22.888168504927307, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.629927396774292, "target_z": 0.6484374562813776, "ce": 6.046441555023193, "genuineness": 4.6956281661987305, "identity": 0.7394976019859314, "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/emo_Pain/dd98b0c40c19f6be6e9357277762a4e3af0fd2521079526eca09459a735552f7.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": "ddf9da767d99a5f873fe9681e4d312a696db48366de6ce1d864a3ea28118b9b2", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Je bericht kwam nadat ik vertrokken was.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik las het twee keer... en legde mijn telefoon weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "5a0ca81d7512df52634d3b43b872282dde6be34e9bf1885a317983ba003b7237", "generation_seconds_batch": 22.25505174405407, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.128709077835083, "target_z": 1.5156250744835789, "ce": 6.125868320465088, "genuineness": 4.086852550506592, "identity": -0.025845160707831383, "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/emo_Pain/ddf9da767d99a5f873fe9681e4d312a696db48366de6ce1d864a3ea28118b9b2.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": "de3a5a6cdcd188e0caa7366a1ecd566d8faa16d021f5f578d81a6107f72a5fa2", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The road turns left behind the bridge.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Keep going... until you reach the small wooden gate.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "338ae73cbfd7e60d5652007454a9489df0db8407a5464c4ff4c7b20c64f2c39e", "generation_seconds_batch": 22.86036586901173, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9265003204345703, "target_z": 1.16406258257962, "ce": 6.006728172302246, "genuineness": 3.8546390533447266, "identity": 0.5053644180297852, "hypothesis": "The road turns left behind the bridge. Keep going until you reach the small wooden gate. We'll meet you on the other side.", "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/emo_Pain/de3a5a6cdcd188e0caa7366a1ecd566d8faa16d021f5f578d81a6107f72a5fa2.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": "de66900d813623257c21165176c804431f4a808d6e704d12d1b891819fce963e", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai écrit le numéro sur l’enveloppe.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lis-le bien... et vérifie les deux derniers chiffres.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "418a12d61e5f5b7ade5fd4a246252fe430d312761c43fc02d028334884306fa6", "generation_seconds_batch": 16.699873942066915, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6703691482543945, "target_z": 0.7187499546621694, "ce": 6.0305562019348145, "genuineness": 4.519899368286133, "identity": 0.8074917793273926, "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/emo_Pain/de66900d813623257c21165176c804431f4a808d6e704d12d1b891819fce963e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "de8906c726870435f6c2a05d60e6490eb65215bc91db5aa5a6157cc300f0aae2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dcabffea87d586eb2f924de9d5bf1bb936d95d358de8b5ecc2c8b28d7b22591e", "generation_seconds_batch": 16.694473252049647, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2635148763656616, "target_z": 1.75, "ce": 5.871703147888184, "genuineness": 4.457139015197754, "identity": 0.6391474604606628, "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/emo_Pain/de8906c726870435f6c2a05d60e6490eb65215bc91db5aa5a6157cc300f0aae2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "df1b2aaf449b1a0eb0ca5ca82bd99241ee21eac502ced349f279ffb58bcf19ce", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c7f1fbaeab700820b1e3446e8dbd81867d5d92c3e8263cd4bea3a3dc3aa65af4", "generation_seconds_batch": 17.143992713070475, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06261950731277466, "target_z": -0.3378906229759897, "ce": 6.046441555023193, "genuineness": 4.7458367347717285, "identity": 0.6521674394607544, "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/emo_Pain/df1b2aaf449b1a0eb0ca5ca82bd99241ee21eac502ced349f279ffb58bcf19ce.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "df8917b40b8ed1e538959bc64bbd4d6693e757ddada163d7f798b33709d31b8d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3361ec893d69dcb3f6f736f91ea50b0db14ffd15bab59a0997f7306c25ccb963", "generation_seconds_batch": 22.651471155928448, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 6.316492557525635, "genuineness": 4.432034492492676, "identity": 0.413488507270813, "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/emo_Pain/df8917b40b8ed1e538959bc64bbd4d6693e757ddada163d7f798b33709d31b8d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "dfd4ea9818e078ecfa0b03068001bbc814d2875de1de605184365c3f6b87fdfc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f3254291246f1e90574f8aff2286eb97b74f227aa0baef18b2dad973176896df", "generation_seconds_batch": 16.411198070039973, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.1712076663970947, "target_z": 3.3281252169739033, "ce": 5.95112943649292, "genuineness": 4.331617832183838, "identity": 0.3211580812931061, "hypothesis": "Die 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/manual-comparison-v1/audio/emo_Pain/dfd4ea9818e078ecfa0b03068001bbc814d2875de1de605184365c3f6b87fdfc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "e0f431dad50d46537829cf7e05b2cdd86bce4e4f8d3ad49e02183f67bbedc20a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "38af22b682b270c3a6630bc8d427db0549eb5f073a57a6c681995965b45be4da", "generation_seconds_batch": 21.23932649404742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 2.0903241634368896, "target_z": 3.18750022021232, "ce": 5.7684478759765625, "genuineness": 4.482243061065674, "identity": 0.7349023818969727, "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/emo_Pain/e0f431dad50d46537829cf7e05b2cdd86bce4e4f8d3ad49e02183f67bbedc20a.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": "e143317540d747cf7d0d0cf715b5c8474ac07cb8889f86a69e2df490c3f0f54d", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Your message arrived after I had left.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Let me know when we can talk.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Your message arrived after I had left.\n(Natural clear speech.) [3.8 seconds duration] I read it twice... before putting the phone away.\n(Natural clear speech.) [3.2 seconds duration] Let me know when we can talk.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "58e855bd334e93ce29d7bddc813dfc1510b9c10c317021f49e223b5d568385e0", "generation_seconds_batch": 18.4809860419482, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.44232261180877686, "target_z": 0.3222656003070744, "ce": 6.419747352600098, "genuineness": 4.394378662109375, "identity": 0.6048157215118408, "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/emo_Pain/e143317540d747cf7d0d0cf715b5c8474ac07cb8889f86a69e2df490c3f0f54d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "e14b509338f46b2f2a2a2a449880638d27376b2bcc96163c2c5f2c7808a1e2b5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "89dc37c5c6bcfa1bdc0f7fe96a8e4bf16607106aa4230e18b563d790f158a5b1", "generation_seconds_batch": 15.748068358981982, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07272994518280029, "target_z": -0.3203124983807918, "ce": 6.157639026641846, "genuineness": 4.7081804275512695, "identity": 0.6865023374557495, "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/emo_Pain/e14b509338f46b2f2a2a2a449880638d27376b2bcc96163c2c5f2c7808a1e2b5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "e1fc12cb62813c0f0e2b53f5d4aef25aae94fb2b9284adbcdb4aa409be13709e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a1dd322b188390c36247dc100d7a0a04ec7fda60da4d8ef36467eda15983573d", "generation_seconds_batch": 23.086049281060696, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3311077952384949, "target_z": 0.12890622975989707, "ce": 5.935244083404541, "genuineness": 4.720732688903809, "identity": 0.7358868718147278, "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/emo_Pain/e1fc12cb62813c0f0e2b53f5d4aef25aae94fb2b9284adbcdb4aa409be13709e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "e203b562cf78e50b29363b7f681afc6f5421039dd004412666689f2d20f7a65f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e6ded07ef5cfb77ea7803eacab88eba3fb5c979cc6fcd3e57965047460ef8af0", "generation_seconds_batch": 16.79077940399293, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.14743484556674957, "target_z": -0.19042969175042163, "ce": 6.070269584655762, "genuineness": 4.65797233581543, "identity": 0.4375678300857544, "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/emo_Pain/e203b562cf78e50b29363b7f681afc6f5421039dd004412666689f2d20f7a65f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "e24fb28ca059c2251ca8d187a9a27b6e86d5cde72e3a3fa6f5b349a1262415d4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "1e619c06ff487f1cef0c1c7e013a81994452a818412d72d4c440c3bb0a67c4c4", "generation_seconds_batch": 17.573600075091235, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4513096809387207, "target_z": 0.3378906229759897, "ce": 5.748591423034668, "genuineness": 4.507347106933594, "identity": 0.6296533942222595, "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/emo_Pain/e24fb28ca059c2251ca8d187a9a27b6e86d5cde72e3a3fa6f5b349a1262415d4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "e2d1184db08a13bcb11cd9a9d022d4078b66c87413cda9a488ccab0fcbd83d2b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f17608228c8c811dbf47e01e60ba13231e944a2bfcf5537652d5c3b2ffbe2ff5", "generation_seconds_batch": 17.76864701299928, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2140860557556152, "target_z": 1.664062478950293, "ce": 5.855817794799805, "genuineness": 4.557555198669434, "identity": 0.6927396059036255, "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/emo_Pain/e2d1184db08a13bcb11cd9a9d022d4078b66c87413cda9a488ccab0fcbd83d2b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "e2e4bc7a92b3d2feb0a9a81e5c4cd4e9352686eef18def2884b5b8cf4c2c525f", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "865c84becee30d4397c4d203617855e14656c6bd93b9900def93b91794053543", "generation_seconds_batch": 19.04886491491925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1642855703830719, "target_z": -0.16113282606086896, "ce": 6.133810997009277, "genuineness": 4.325342178344727, "identity": 0.4099583327770233, "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/emo_Pain/e2e4bc7a92b3d2feb0a9a81e5c4cd4e9352686eef18def2884b5b8cf4c2c525f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "e39dcc901def4b42266d01adc4327e4b28df5041a14e1e9c7d92f95a978f096b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "9f971329d24698ef0e7df04048de106b46ab07e8fe8f1d9129d5f6298eef4908", "generation_seconds_batch": 22.777452719048597, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 6.181467056274414, "genuineness": 4.024092197418213, "identity": 0.14005087316036224, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/e39dcc901def4b42266d01adc4327e4b28df5041a14e1e9c7d92f95a978f096b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "e411bba2dd79ceab33e54bbd5aaf4ed4871ffa5a93ff36c58ff4cdd60a9d578c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "55ae138885d09257e184c9fbf377224b83369920e5b6483b41a4994a4e69cf89", "generation_seconds_batch": 23.972587067983113, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5467971563339233, "target_z": 0.5039062556672288, "ce": 5.804190158843994, "genuineness": 4.7332844734191895, "identity": 0.6462457180023193, "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/emo_Pain/e411bba2dd79ceab33e54bbd5aaf4ed4871ffa5a93ff36c58ff4cdd60a9d578c.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": "e4799a9840ce0bca29a5093d0bad92cea7a3ae7a4adaf0531a8a7ca483635fdd", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ton message est arrivé après mon départ.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Je l’ai relu... puis j’ai rangé mon téléphone.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c5321026d837be29948fcf5b93904d2b427a42cce36b89d15cc2ff85bdcc43df", "generation_seconds_batch": 16.8346429169178, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1826313734054565, "target_z": 1.6093750032384164, "ce": 5.927301406860352, "genuineness": 4.444586753845215, "identity": 0.7230675220489502, "hypothesis": "Ton message est arrivé après mon départ. Je l'ai relu, puis j'ai 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/emo_Pain/e4799a9840ce0bca29a5093d0bad92cea7a3ae7a4adaf0531a8a7ca483635fdd.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "e4e7407fd8d3bf87f24759ccfdc6c7ffb6ecca9b4ce73c96b525a80a98367a1e", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c90f6ef4339cb88d96a444299c7b1510c729be46fe16bf0a452c083b21306a00", "generation_seconds_batch": 16.82599448703695, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11373338103294373, "target_z": -0.24902344903685866, "ce": 6.213237762451172, "genuineness": 4.582659721374512, "identity": 0.32327210903167725, "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/emo_Pain/e4e7407fd8d3bf87f24759ccfdc6c7ffb6ecca9b4ce73c96b525a80a98367a1e.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": "e50f70c0e266a126aa286b1ea0c6f4e49f670619b9849320892b3ef4db0408ed", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Te esperaré al otro lado.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Natural clear speech.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Natural clear speech.) [3.2 seconds duration] Te esperaré al otro lado.", "seed": 1701, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "73e9aace8b78a468634c694396b5cd635c74cfc39f0287afa76146b38282c726", "generation_seconds_batch": 16.941421539057046, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20725494623184204, "target_z": -0.08642577062394596, "ce": 5.816103935241699, "genuineness": 4.394378662109375, "identity": 0.6694523096084595, "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/emo_Pain/e50f70c0e266a126aa286b1ea0c6f4e49f670619b9849320892b3ef4db0408ed.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "e59c99e44d2193cac84b3466369a4f6d98f2b0b9a29dd5ab08e345c59ede7f33", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3ea6b0584425666986b736c662386ce864665f315573c10d8031ed612608eccc", "generation_seconds_batch": 23.45004660298582, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08621053397655487, "target_z": -0.2968749902847506, "ce": 6.24500846862793, "genuineness": 4.507347106933594, "identity": 0.6437621712684631, "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/emo_Pain/e59c99e44d2193cac84b3466369a4f6d98f2b0b9a29dd5ab08e345c59ede7f33.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": "e59cb482b6a497866223728fb8c5d092179fddfd117954e314534f33594dc9ef", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La carretera gira a la izquierda tras el puente.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sigue... hasta la pequeña puerta de madera.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "23cb3e07eeba1870aa42cdb847f7e967ef605126a89da0f0d6484becf581d321", "generation_seconds_batch": 21.07561269099824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1152284145355225, "target_z": 1.492187436850879, "ce": 5.974957466125488, "genuineness": 4.218649387359619, "identity": 0.6243033409118652, "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/emo_Pain/e59cb482b6a497866223728fb8c5d092179fddfd117954e314534f33594dc9ef.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": "e5a05ba379da76b674811fdfda997eecd77f43aeea15fe222ea5f2c8447016c5", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Schlüssel liegen auf dem Küchentisch.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich suchte in der Schublade... dann neben dem Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "631eb4bb9f8aec2358357b958d0bdc5d2ad6dfcd9b2c4f720b473b0e742bd136", "generation_seconds_batch": 17.369245629990473, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28280237317085266, "target_z": 0.044921862451136184, "ce": 5.816103935241699, "genuineness": 4.319066047668457, "identity": 0.6554317474365234, "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/emo_Pain/e5a05ba379da76b674811fdfda997eecd77f43aeea15fe222ea5f2c8447016c5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "e63dd026311cdd85a6b289e1a6cecae0823426259de3093fe16c7c1850d62b0e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "c1d1a03271f440b0d98158660acb7bbcdc9d63860da126a6d9b09c88374e094a", "generation_seconds_batch": 30.14840353198815, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38783860206604004, "target_z": 0.22753906877443192, "ce": 5.720792293548584, "genuineness": 3.986435890197754, "identity": 0.4636831283569336, "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/emo_Pain/e63dd026311cdd85a6b289e1a6cecae0823426259de3093fe16c7c1850d62b0e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "e6a99b93aeb47a44505a0660dfeec9d8044a265a09ebb6ee0eae812ee35f674e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5bf9bd2246f94664af96c3c46494d324118590b31d116cf8c6f14669874e00ba", "generation_seconds_batch": 22.945940076955594, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10980154573917389, "target_z": -0.25585938350084325, "ce": 5.8875885009765625, "genuineness": 4.557555198669434, "identity": 0.6095543503761292, "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/emo_Pain/e6a99b93aeb47a44505a0660dfeec9d8044a265a09ebb6ee0eae812ee35f674e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "e6d5af87a018058c55a0da66c5865577e22730378bbda41841a12b7deade3294", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f1dc915411345ce0974b27e995176c96e4350be329e45990ce632f2b0184e88f", "generation_seconds_batch": 17.66364781104494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1030612587928772, "target_z": -0.2675781245951979, "ce": 6.395919322967529, "genuineness": 4.758388519287109, "identity": 0.2769128978252411, "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/emo_Pain/e6d5af87a018058c55a0da66c5865577e22730378bbda41841a12b7deade3294.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "a84a459e8d7d225d24de55cba87569472d60eb887c94fe149c095cca463a1fa3", "generation_seconds_batch": 17.42886963998899, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.268836498260498, "genuineness": 4.457139015197754, "identity": 0.3977854251861572, "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": 0.03999999999999915, "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/emo_Pain/e6f1e80cc6606ef00fdc322d5829b94fd65ad80a74ab015f88321e5dc194804c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "e7c91c3f5fa1c79dcb275fe3e201c41cda5b5d8c53985849757c20024d2141ef", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "94256a43205241d19949e21812c3de7f97ffc2d1d23b6d6fea28870c41af0d5e", "generation_seconds_batch": 22.850971462088637, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 5.927301406860352, "genuineness": 4.632867813110352, "identity": 0.6840249300003052, "hypothesis": "Te he 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/emo_Pain/e7c91c3f5fa1c79dcb275fe3e201c41cda5b5d8c53985849757c20024d2141ef.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "e7cb63c08ab5c4cd9840c8dee7fe23d2fd895f42ffe30f9a59d5a40b8e9836bf", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8143953ecaf52e36af17ee43a9fd41b0cd9c25e7200637761b70cc0c36bd7260", "generation_seconds_batch": 22.226331147016026, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11317169666290283, "target_z": -0.25, "ce": 6.038498878479004, "genuineness": 4.20609712600708, "identity": 0.23770108819007874, "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/emo_Pain/e7cb63c08ab5c4cd9840c8dee7fe23d2fd895f42ffe30f9a59d5a40b8e9836bf.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "e7cc1f371bc7c02fedf1d6b4a12f14f799b2b56339caff097e54f36dd6322a12", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "090f03601c57be1aa88ca79919b33e0067955c2c99c36921795113e7333d7281", "generation_seconds_batch": 18.30679852794856, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9939031600952148, "target_z": 1.2812499417085035, "ce": 5.6532793045043945, "genuineness": 4.306513786315918, "identity": 0.5188911557197571, "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/emo_Pain/e7cc1f371bc7c02fedf1d6b4a12f14f799b2b56339caff097e54f36dd6322a12.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "e7ed4713c3bc64e07449660ae729281709b63da7631386487b5182122b34016e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "34bdbbf2b291ab725a47323e66ce81bfbeb8ce8204d44f9eba54cfb585d7d2d1", "generation_seconds_batch": 23.9398294739658, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1511766910552979, "target_z": 1.55468752752654, "ce": 5.474569320678711, "genuineness": 4.061748504638672, "identity": 0.42663896083831787, "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/emo_Pain/e7ed4713c3bc64e07449660ae729281709b63da7631386487b5182122b34016e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "e810328efe5e15c42410843e13370bec4c61a704aa125e3ff8a65a1f8a30e72b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "74f9f77ea02b8a3e025dafc18b9a22bd718823dc7dfb3f8e78c2c562717f410e", "generation_seconds_batch": 17.51046027895063, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6703691482543945, "target_z": 0.7187499546621694, "ce": 5.522225379943848, "genuineness": 4.457139015197754, "identity": 0.5992786884307861, "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/emo_Pain/e810328efe5e15c42410843e13370bec4c61a704aa125e3ff8a65a1f8a30e72b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "e82944b221050a78be559404f1cd400ea114ea6c157a060ad021d2757e6e092d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7489e112a294235f260c628126d862539d9ed20394aaa3dcdae00ba617e0b310", "generation_seconds_batch": 22.140574865974486, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.20809747278690338, "target_z": -0.08496094417923397, "ce": 6.078212261199951, "genuineness": 4.519899368286133, "identity": 0.6029748916625977, "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/emo_Pain/e82944b221050a78be559404f1cd400ea114ea6c157a060ad021d2757e6e092d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "e8b385dcfe2ee35f57c0b74766c21388be95d0c191e6225f9bf479adeac21e66", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7625ed945cf7f5213aa9bb7d35a58e7c75d32cc84e78b93a457ac8916681dc6a", "generation_seconds_batch": 17.66364781104494, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.44648033380508423, "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/emo_Pain/e8b385dcfe2ee35f57c0b74766c21388be95d0c191e6225f9bf479adeac21e66.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": "e8b9d8ab3c141934f7a603ef321f135951b2035caf891b3f57e0f70c85a02710", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Quelqu’un a déplacé la petite chaise bleue.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était ici... près de la grande bibliothèque.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 128, "seconds_generated": 10.24, "mp3_sha256": "70a2943602034c79d742fa70acc3ce1832fe9ac394bc0ad0a071f4f74070b9c6", "generation_seconds_batch": 17.530481061083265, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3895236849784851, "target_z": 0.2304687734785194, "ce": 6.006728172302246, "genuineness": 4.231201171875, "identity": 0.8023145198822021, "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/emo_Pain/e8b9d8ab3c141934f7a603ef321f135951b2035caf891b3f57e0f70c85a02710.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "e9225f29331a8dadd6cb496ba2037b0b5caaa8c84583d9b60b50c0cc6e018d09", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e8af7d62134b3117866a93423c650f07e0b38189ec188afb516aaa75c3f7ecdd", "generation_seconds_batch": 23.208367681945674, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07610009610652924, "target_z": -0.31445311487994854, "ce": 6.260893821716309, "genuineness": 4.18099308013916, "identity": 0.49920833110809326, "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/emo_Pain/e9225f29331a8dadd6cb496ba2037b0b5caaa8c84583d9b60b50c0cc6e018d09.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "e95c9619edf591c86d869ad021aa0a9d898b3c7dacb69d01b5d8d217b6aee374", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dc1a05fb483e88b608c7fa00c34c2d8875dcbaba9fa9c8e2889a7dc52d31a33b", "generation_seconds_batch": 22.454918506089598, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.04689215123653412, "target_z": -0.3652343867392597, "ce": 5.95112943649292, "genuineness": 4.287685871124268, "identity": 0.5134695768356323, "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/emo_Pain/e95c9619edf591c86d869ad021aa0a9d898b3c7dacb69d01b5d8d217b6aee374.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": "e96c24821f646b1d5d9d58dfc8e3e1685a4f61183b8a2c65130ac3df3f1fdc9b", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Someone moved the little blue chair.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] We can put it back after lunch.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Someone moved the little blue chair.\n(Natural clear speech.) [3.8 seconds duration] It used to stand here... beside the tall bookshelf.\n(Natural clear speech.) [3.2 seconds duration] We can put it back after lunch.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e19da81bfe62accd9024d800da411c2d1288a77858f3a917ce75453ed3df8dff", "generation_seconds_batch": 17.879899521009065, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13114580512046814, "target_z": -0.21875000647683293, "ce": 6.109982967376709, "genuineness": 4.607763767242432, "identity": 0.35889044404029846, "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/emo_Pain/e96c24821f646b1d5d9d58dfc8e3e1685a4f61183b8a2c65130ac3df3f1fdc9b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "e9c9a1c70e72d960d907613d4b6d5dd28ff9e5aea08b4c00c8a3804e065ce045", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e91652ba6cff65af183702bbc8dfa4381f0288eb259b3f4658c005221b674f09", "generation_seconds_batch": 16.78667964693159, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08845728635787964, "target_z": -0.29296876052485354, "ce": 6.205295085906982, "genuineness": 4.7458367347717285, "identity": 0.4173298478126526, "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/emo_Pain/e9c9a1c70e72d960d907613d4b6d5dd28ff9e5aea08b4c00c8a3804e065ce045.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "a17c6049fb0a34df2dbc42fce33ee2bb8d2fcc550ccdea3aa4138da965d85a74", "generation_seconds_batch": 17.549043214996345, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.523001670837402, "genuineness": 4.65797233581543, "identity": 0.5385887622833252, "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/emo_Pain/e9d266c3907611a53d0c5f7688a7e231bba29ead6f87b8ae8d29cb37ab2fae41.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "ea100d208b70ec78419df7e200b21928d0c8c8d473a8ab21d332ebff34ca824b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] He encontrado la cajita.\n(Natural clear speech.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Natural clear speech.) [3.2 seconds duration] Podemos mirar juntos lo que hay dentro.", "seed": 1702, "scale": 0.75, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "f618c17a05bb222784a9a0246df639c0974e09c104241f09b8ebbef569077ae2", "generation_seconds_batch": 22.21788643707987, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6209403276443481, "target_z": 0.6328124336124624, "ce": 5.550024509429932, "genuineness": 4.193545341491699, "identity": 0.5460180044174194, "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/emo_Pain/ea100d208b70ec78419df7e200b21928d0c8c8d473a8ab21d332ebff34ca824b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "ea3e1f2e83b43c8fe594855cd194e40871f491fcc42ffa39acb5a930f1f5756d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7ff526c1ca1c83252857fe2bfb13cd2da877a48d1dc5e20869bac80ddc8a2882", "generation_seconds_batch": 22.277494769077748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3174371719360352, "target_z": 1.8437499287548378, "ce": 5.4666266441345215, "genuineness": 4.231201171875, "identity": 0.25433990359306335, "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/emo_Pain/ea3e1f2e83b43c8fe594855cd194e40871f491fcc42ffa39acb5a930f1f5756d.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": "ea482def9ec5e594f882e5ad89efcfcf41dcb5548b0008f381a262b6c076c6a3", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich schrieb die Nummer auf den Umschlag.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lies sie genau... und prüfe die letzten beiden Ziffern.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "93566983002524bf176ade88fec79692ab4d17986aecea0d9129b4060f42b6f8", "generation_seconds_batch": 17.241347386036068, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.23618203401565552, "target_z": -0.03613280015353721, "ce": 5.855817794799805, "genuineness": 4.532451152801514, "identity": 0.8014618158340454, "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/emo_Pain/ea482def9ec5e594f882e5ad89efcfcf41dcb5548b0008f381a262b6c076c6a3.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "ea8a1293d718346470c059f4268969c2d99e461e8799ef84f19be16f3ff1f0db", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "39ce8e91cdbbebe0bd9bead77a4574f7a8ac163ffe48609df2aa63f6c2bb5a18", "generation_seconds_batch": 16.336989733972587, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12328213453292847, "target_z": -0.23242187540480205, "ce": 6.23706579208374, "genuineness": 4.833701133728027, "identity": 0.5021423101425171, "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/emo_Pain/ea8a1293d718346470c059f4268969c2d99e461e8799ef84f19be16f3ff1f0db.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "eac7b1b53b464de690e2ab1243c548d3f74cc19740850c0a3db33f8544ef2373", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "92b596ec73de3e83b6bc53a195c249714d55eea384034f71bb6390f26111c1ef", "generation_seconds_batch": 23.447647182969376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4097445607185364, "target_z": 0.2656250226689153, "ce": 5.569881439208984, "genuineness": 4.607763767242432, "identity": 0.6075441837310791, "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/emo_Pain/eac7b1b53b464de690e2ab1243c548d3f74cc19740850c0a3db33f8544ef2373.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "ead3bcf4393c65ef758d990fdea14cd8894db0eb87b706665f1fa2bccff88910", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "21ee0c64b18b89ce791fcd70f789b68780bebee65e3ed512932b3a2b15bfede3", "generation_seconds_batch": 27.145551898982376, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.39907240867614746, "target_z": 0.2470702952959125, "ce": 5.855817794799805, "genuineness": 4.557555198669434, "identity": 0.4122489392757416, "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/emo_Pain/ead3bcf4393c65ef758d990fdea14cd8894db0eb87b706665f1fa2bccff88910.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "eafce140c5997c92d781864a69664ca06c6c0b87a72bd6baac878761f95ccd14", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4e51b2517155dde8c6379e664a2bfb1b990b0037a36f9d40df0e4a064dffb2a9", "generation_seconds_batch": 18.678087811917067, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.015437468886375427, "target_z": -0.41992186245113616, "ce": 5.95112943649292, "genuineness": 4.432034492492676, "identity": 0.6971743106842041, "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/emo_Pain/eafce140c5997c92d781864a69664ca06c6c0b87a72bd6baac878761f95ccd14.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "eb1eca04bd0b9235bd0e5c8d994f16eb0f088d1ba768c85b317474d238850015", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "4c69045a24419258bea794cb8d37179e3be02ba7e28df09feb6bf6a84feac3fa", "generation_seconds_batch": 17.55427448498085, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.334477961063385, "target_z": 0.13476563916807205, "ce": 6.125868320465088, "genuineness": 4.3504462242126465, "identity": 0.4947722256183624, "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/emo_Pain/eb1eca04bd0b9235bd0e5c8d994f16eb0f088d1ba768c85b317474d238850015.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "eb240ac3e148646d8cf1d62882bc8fcf5dcd409b7bd399bfdbb0af15f2a3713b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d85706df23564c0618d061f8c01a4ed07efefe307a0a667f9521caed6c47dcf5", "generation_seconds_batch": 29.939873318071477, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11598014831542969, "target_z": -0.24511719336962984, "ce": 6.316492557525635, "genuineness": 4.3881025314331055, "identity": 0.16281592845916748, "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/emo_Pain/eb240ac3e148646d8cf1d62882bc8fcf5dcd409b7bd399bfdbb0af15f2a3713b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "ebaaa44c90c2bd9eda1e63f238a2a8b4650d0cc4421ef352093ba2fc9a1c7966", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b14f45004a2d01f4a4f50e736942bbc6b166782daadcb5223d67cfa456badb3d", "generation_seconds_batch": 22.3744133800501, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.3367246985435486, "target_z": 0.13867184302063737, "ce": 5.863759994506836, "genuineness": 4.507347106933594, "identity": 0.5067133903503418, "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/emo_Pain/ebaaa44c90c2bd9eda1e63f238a2a8b4650d0cc4421ef352093ba2fc9a1c7966.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "83426dae2ed3fa139e6bcea3b2ff35a203b3a4eda48110589c0edda20955dedf", "generation_seconds_batch": 17.164145344984718, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.062326908111572, "genuineness": 4.344170093536377, "identity": 0.7704787254333496, "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/emo_Pain/ebe89981a89e9506d02bacdfa583f9bfe1f04faa699193749ab6419de85c588f.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "eca1f59c69a5d04f0213c13f519891f03231f6856c9fa560ad3707acd08b5472", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f875144f1e4be5c7cc43329c0a4d633c3d040a1842e8a0574cfbea50445745df", "generation_seconds_batch": 17.75086737505626, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5939792394638062, "target_z": 0.5859375728643705, "ce": 5.661221981048584, "genuineness": 4.632867813110352, "identity": 0.7069962620735168, "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/emo_Pain/eca1f59c69a5d04f0213c13f519891f03231f6856c9fa560ad3707acd08b5472.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "ed1756f8a4795f276c3c9139ce4721f5320b2dfe67847a9cc9715c31b48d0a29", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.5, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "18020308f435afad6580edeec642dc32285aa088c167b81a3c01cf50ed0cef56", "generation_seconds_batch": 21.23932649404742, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5535374879837036, "target_z": 0.5156250744835787, "ce": 4.939431667327881, "genuineness": 4.356721878051758, "identity": 0.6489956974983215, "hypothesis": "La porte... La porte 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/emo_Pain/ed1756f8a4795f276c3c9139ce4721f5320b2dfe67847a9cc9715c31b48d0a29.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "ed60ab10eaafebb6976e48f6824c787c5064f33e4d871905cd454af382454845", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e830db29dd3dc5c979fc1eb8985fcbb3aad85d54f57483a4dbdb918daacacd39", "generation_seconds_batch": 24.032599430996925, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6366677284240723, "target_z": 0.6601562750977277, "ce": 5.450741291046143, "genuineness": 4.419482707977295, "identity": 0.2714284062385559, "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/emo_Pain/ed60ab10eaafebb6976e48f6824c787c5064f33e4d871905cd454af382454845.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "ed7e92b31efe736f851a778ed45f2769f3ebb58071211aaeed8e89a2f805f25b", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "5036b0eaae0f20daccf4caf293682440e698dd78b41a8a435269a1f45bf222ed", "generation_seconds_batch": 22.277494769077748, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1242154836654663, "target_z": 1.5078124595197941, "ce": 5.4666266441345215, "genuineness": 4.469690799713135, "identity": 0.6795692443847656, "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/emo_Pain/ed7e92b31efe736f851a778ed45f2769f3ebb58071211aaeed8e89a2f805f25b.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "ed8aa1e3b642d521c9b39b223281ffac2202004825db8fd32bb771e775dfcd0e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "8f85ed4d1ba9e111c84b1c8bda0a42691fadfda0772085d5c4c5a11686ce0cde", "generation_seconds_batch": 21.70856671698857, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.7827073335647583, "target_z": 0.9140624271356295, "ce": 5.7684478759765625, "genuineness": 4.3127899169921875, "identity": 0.39568600058555603, "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/emo_Pain/ed8aa1e3b642d521c9b39b223281ffac2202004825db8fd32bb771e775dfcd0e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "ed92e23854776384eb69b912c5fa583ac3734776cf338d9c71924a224cbb3c81", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6fca89de47d874514fa560a429528a0e5e0bc61bb132f0374a6be7d8169854e1", "generation_seconds_batch": 22.3744133800501, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.15473683178424835, "target_z": -0.1777343737855938, "ce": 5.661221981048584, "genuineness": 4.545003414154053, "identity": 0.5032999515533447, "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/emo_Pain/ed92e23854776384eb69b912c5fa583ac3734776cf338d9c71924a224cbb3c81.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "eda277b7d7536290f015aa756ac6f577bc3799595ef3a1a94b9126173b040e71", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "065999254291615435e7e4fde86b9201deeceb2c4c5af11372708d18c219bbf6", "generation_seconds_batch": 24.395085487980396, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2627219259738922, "target_z": 0.010009763955191509, "ce": 5.720792293548584, "genuineness": 4.6705241203308105, "identity": 0.7067378163337708, "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/emo_Pain/eda277b7d7536290f015aa756ac6f577bc3799595ef3a1a94b9126173b040e71.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "edf31d5f37b11ca09c93f1832c27d49ffd55a1e610244194d033ebfdf89c0f35", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "a7dd1566049552c8ab8d990478fc11cb20654130ad01526a1a558bea2b7c982b", "generation_seconds_batch": 16.381978317978792, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.205295085906982, "genuineness": 4.808597087860107, "identity": 0.7642284035682678, "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/emo_Pain/edf31d5f37b11ca09c93f1832c27d49ffd55a1e610244194d033ebfdf89c0f35.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": "ee1d6733955254b88b6abb70c28b333be17e7f268fa820675755b55a6f0b73fe", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La route tourne à gauche après le pont.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Continue... jusqu’à la petite barrière en bois.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": 127, "seconds_generated": 10.16, "mp3_sha256": "c0f0f7f04f6d468734c5f1aa50636e718148f46c10ce7d3d46dc1a6e36399dbd", "generation_seconds_batch": 23.498963849968277, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2410472631454468, "target_z": 1.7109375469570387, "ce": 5.240260601043701, "genuineness": 4.024092197418213, "identity": 0.599611759185791, "hypothesis": "La route tourne à gauche après le pont. Continue jusqu'à la petite barrière en bois. Je t'attendrai de l'autre côté.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/ee1d6733955254b88b6abb70c28b333be17e7f268fa820675755b55a6f0b73fe.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "eede8cfcc7b2d312bce8bdea7f6e78d867f0ca35a3e02c2305066482e835f518", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "6b5de3c267737e0cd10d16f7307030e97ada97f6dc19849fc69601a45c328554", "generation_seconds_batch": 21.53002063499298, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07272994518280029, "target_z": -0.3203124983807918, "ce": 6.197352409362793, "genuineness": 4.6705241203308105, "identity": 0.7596575021743774, "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/emo_Pain/eede8cfcc7b2d312bce8bdea7f6e78d867f0ca35a3e02c2305066482e835f518.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "ef71c1c11ae72de49b71bd5316bf282c4aeeb4477d2d8c9700362230bc0354d6", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "001fd7b7a27b636a717aaee18fc9807274274481ca63f34c49540a73492b9ab8", "generation_seconds_batch": 22.579880341072567, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10755479335784912, "target_z": -0.2597656132607403, "ce": 6.316492557525635, "genuineness": 4.65797233581543, "identity": 0.462913453578949, "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/emo_Pain/ef71c1c11ae72de49b71bd5316bf282c4aeeb4477d2d8c9700362230bc0354d6.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "efb8681dbe23e66e18a87bed166c947662211bf03841225ea62442406ea3b089", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "3f9e517e19842905217981bff2ec6f2f14eef81a00779a971120c40ebcdfeba0", "generation_seconds_batch": 22.881598870968446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4771474599838257, "target_z": 0.38281248542712587, "ce": 6.181467056274414, "genuineness": 4.6956281661987305, "identity": 0.601325273513794, "hypothesis": "The door is still open. I'd put your coat on the chair. Tell me when you are ready.", "wer": 0.05555555555555555, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/efb8681dbe23e66e18a87bed166c947662211bf03841225ea62442406ea3b089.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "efce50fbf085cd35d05c4201a334184a4cec07f25b49637ff5b5b244c527aeb0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f2b26c4f5900c7639a2af4e7df9fbbd6816204811186cd3729577b9f813a0351", "generation_seconds_batch": 21.88280551705975, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 5.959072113037109, "genuineness": 4.582659721374512, "identity": 0.8027361631393433, "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/emo_Pain/efce50fbf085cd35d05c4201a334184a4cec07f25b49637ff5b5b244c527aeb0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "f02c110500b5fd9304891b795a383e876020bf321180d645e19c43329061d4bd", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 1.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "1e90a7178e35cf34d7afa22b647b738195f3b7718d60b5c111c0bb90245cb85c", "generation_seconds_batch": 24.21055451501161, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1242154836654663, "target_z": 1.5078124595197941, "ce": 5.824047088623047, "genuineness": 4.720732688903809, "identity": 0.7624011039733887, "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/emo_Pain/f02c110500b5fd9304891b795a383e876020bf321180d645e19c43329061d4bd.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "f043235bffb336e41a55663036e520e77167b15d8b8e7c6508991f4fc4373db7", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8d3705f2ce104c969c0b2de4c9fde47c34ecff5f45b919a737b6a9d96ff49dc9", "generation_seconds_batch": 22.54574990295805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2854704260826111, "target_z": 0.04956057020171863, "ce": 6.117925643920898, "genuineness": 4.582659721374512, "identity": 0.5871353149414062, "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/emo_Pain/f043235bffb336e41a55663036e520e77167b15d8b8e7c6508991f4fc4373db7.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": "f0444e477811a98e9090a054a3ee50096c8709974a46e02f1a49c290f53abff1", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Deine Nachricht kam, als ich schon weg war.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich las sie zweimal... dann legte ich das Handy weg.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "1f33a8823abb45642d8fccb3734d9c0fd1e03a556d5e4e578d56e90caf7eb23f", "generation_seconds_batch": 17.364065509056672, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.1691508293151855, "target_z": 1.5859375728643705, "ce": 5.363371849060059, "genuineness": 4.406930446624756, "identity": 0.7332139611244202, "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/emo_Pain/f0444e477811a98e9090a054a3ee50096c8709974a46e02f1a49c290f53abff1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "f0cdc720acb20f4f90645e43c3fee2b674d8c0eedbdb67256bef34d98f7c0604", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "b1376f2804d66bfc6edd69649f4c112a5779fa2bc7266cf3bd67e1871eb0f091", "generation_seconds_batch": 23.04837839002721, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6501482725143433, "target_z": 0.6835937054717736, "ce": 5.589737892150879, "genuineness": 4.545003414154053, "identity": 0.48803383111953735, "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/emo_Pain/f0cdc720acb20f4f90645e43c3fee2b674d8c0eedbdb67256bef34d98f7c0604.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 127, "seconds_generated": 10.16, "mp3_sha256": "bae21ccd77a36f0ca35103c6593e8ef004b1d2e896d68bafb0a8df3a4d7ed08d", "generation_seconds_batch": 18.21413619199302, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0963209718465805, "target_z": -0.27929686568955264, "ce": 6.157639026641846, "genuineness": 4.7458367347717285, "identity": 0.7149426937103271, "hypothesis": "Ton message est arrivé après mon départ, je l'ai relu, puis j'ai rangé mon téléphone. Dis-moi quand nous pourrons parler.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/f21bc0adf2df6aeda4aa5bc68657c60b57559ffed3b647118e970ebfd0ba6ad8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "f32e8588042919b0f0b5bea372fa8e57d9c58a503b9d3b2c3d1073175d08c16c", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo estás listo.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La puerta sigue abierta.\n(Natural clear speech.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo estás listo.", "seed": 1701, "scale": 1.0, "cfg": 1.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "57e3b87ec742881a070061e639b9e38a5915e6ff5c1cdc035dd79b1f0e3b71fc", "generation_seconds_batch": 21.29163441399578, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.38615351915359497, "target_z": 0.2246093640703444, "ce": 5.982900142669678, "genuineness": 4.68307638168335, "identity": 0.6877727508544922, "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/emo_Pain/f32e8588042919b0f0b5bea372fa8e57d9c58a503b9d3b2c3d1073175d08c16c.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "f38e272e587d224644cb9518cfd1c85d1e889c6fdf2fffdc90b21549330dc0d5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "0325645cfb8f3be34bc705c2bac135932d1f9af89b45bb858b736b2aef9f0a79", "generation_seconds_batch": 21.688529993058182, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10867816209793091, "target_z": -0.25781251133445765, "ce": 6.324435234069824, "genuineness": 4.344170093536377, "identity": 0.6031398773193359, "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/emo_Pain/f38e272e587d224644cb9518cfd1c85d1e889c6fdf2fffdc90b21549330dc0d5.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "f4ca11ff5c396dbaac4745c3b20ff9a0505cee50775474e34b0640d7425a0691", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "191b1d7e7dc5c6f67ef37facb7a7d9e767ad872a90c332f132e5a7438e9aeba6", "generation_seconds_batch": 18.72962139698211, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059249356389045715, "target_z": -0.34375000647683296, "ce": 6.062326908111572, "genuineness": 4.13078498840332, "identity": 0.6270419359207153, "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/emo_Pain/f4ca11ff5c396dbaac4745c3b20ff9a0505cee50775474e34b0640d7425a0691.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "f539d22f0ca32039b64ec22b557c727b50a3fc3f1095b8917ae84952f6f73a9d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ce966236fcb5fb2bc4796bd5d52fe8386551643f331f48c46987c8aeaf0e5694", "generation_seconds_batch": 23.056892651016824, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.3848401308059692, "target_z": 1.9609374951423753, "ce": 5.708878040313721, "genuineness": 4.168440818786621, "identity": 0.16720834374427795, "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/emo_Pain/f539d22f0ca32039b64ec22b557c727b50a3fc3f1095b8917ae84952f6f73a9d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "f5c0aacefdad16f6a98e2ef7cd77fc2db48924488799f597148ad8dbb06e5b24", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7f48424269cf3540f5275d8a16a4a8af58d852e1633d98a8e5f7b0156f08e7c1", "generation_seconds_batch": 22.881598870968446, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.10193787515163422, "target_z": -0.2695312524288124, "ce": 6.149696350097656, "genuineness": 4.293961524963379, "identity": 0.4771977663040161, "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/emo_Pain/f5c0aacefdad16f6a98e2ef7cd77fc2db48924488799f597148ad8dbb06e5b24.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "f5c3e1a7c4a3e34e7a323144ad78e34d62b1f862f6b263bec1086d86a8ad16b4", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1702, "scale": 0.75, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "e6ca7cfafaf6de4b56136638ec3852022e69f82544c3faac1bab866d35ee5888", "generation_seconds_batch": 23.373129783081822, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.6434080004692078, "target_z": 0.6718749902847506, "ce": 5.7684478759765625, "genuineness": 4.519899368286133, "identity": 0.6864204406738281, "hypothesis": "La porte, t'as encore ouverte, 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/manual-comparison-v1/audio/emo_Pain/f5c3e1a7c4a3e34e7a323144ad78e34d62b1f862f6b263bec1086d86a8ad16b4.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "f62b76a92b4d25acdea204d816941573f89e74860c694820d4835a616ec4e32e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "18ffcc112ac951e8921efaef9b2185625eed1117fa9ae1c2e2f0b36353207cc3", "generation_seconds_batch": 24.41671281994786, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.28280237317085266, "target_z": 0.044921862451136184, "ce": 5.816103935241699, "genuineness": 4.570107460021973, "identity": 0.6269100904464722, "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/emo_Pain/f62b76a92b4d25acdea204d816941573f89e74860c694820d4835a616ec4e32e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "f6fa5ffbd799d65762c271c0f0aabb953b9454222eeb3d8218a5ecb1c241446e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ae8743e396b681e374c52e7c0293158b7866285ec44f9b63c1314ec1f6650503", "generation_seconds_batch": 23.636065913946368, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.7937511205673218, "target_z": 2.6718748866554236, "ce": 5.879645347595215, "genuineness": 4.344170093536377, "identity": 0.09055006504058838, "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/emo_Pain/f6fa5ffbd799d65762c271c0f0aabb953b9454222eeb3d8218a5ecb1c241446e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "f6fd20e5b54b4a810d3d454e1681ff87f49ab425230f075898bc1bfcc5530234", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "7ed00158f561a6138ca90143a34ab250cba0a862b863188764b5950ee39b2281", "generation_seconds_batch": 22.3828182220459, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06711302697658539, "target_z": -0.33007813754886384, "ce": 6.125868320465088, "genuineness": 4.984326362609863, "identity": 0.7316012382507324, "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/emo_Pain/f6fd20e5b54b4a810d3d454e1681ff87f49ab425230f075898bc1bfcc5530234.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "f70e043fae0583e74ae7cfdfdc4a190f462bf750de49bdc787329d20815a40a2", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d4c67ce62795bccaf2b880362464ac06031de184fc5eb5fe0ddc35d3f84d8d2a", "generation_seconds_batch": 23.210424395976588, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.957955002784729, "target_z": 1.2187500582914965, "ce": 4.9900665283203125, "genuineness": 4.532451152801514, "identity": 0.5233197808265686, "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/emo_Pain/f70e043fae0583e74ae7cfdfdc4a190f462bf750de49bdc787329d20815a40a2.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "f7441c9a3cae1f0ea132e812031c95dfa58e08b0e3ed0ade090a2fb7e0550922", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8dff4f11315d07dc3af00d58a8b01dd3707ec3e89758db507778c784351c6da5", "generation_seconds_batch": 22.329000617028214, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.22466737031936646, "target_z": -0.056152328063920225, "ce": 5.903473377227783, "genuineness": 4.7834930419921875, "identity": 0.7897879481315613, "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/emo_Pain/f7441c9a3cae1f0ea132e812031c95dfa58e08b0e3ed0ade090a2fb7e0550922.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-1", "key": "f7769cdc1fca3683b620247a26767f7638491400a1c00250bc03583757ed50c5", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ich habe die kleine Schachtel gefunden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Sie stand neben dem alten Fenster.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d55fabd96f29a2ca9ca2a572ecb38a9a600e0ae2c681ecbe84b6bfff856b5572", "generation_seconds_batch": 23.587319644051604, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0792802572250366, "target_z": 1.4296875534338718, "ce": 5.919358730316162, "genuineness": 4.595211505889893, "identity": 0.7546589374542236, "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/emo_Pain/f7769cdc1fca3683b620247a26767f7638491400a1c00250bc03583757ed50c5.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "1b80b24b257e217c9983b05a468716846f26d853562f74cba25db30b784c9e08", "generation_seconds_batch": 16.531670378055423, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07610009610652924, "target_z": -0.31445311487994854, "ce": 6.165581703186035, "genuineness": 4.6956281661987305, "identity": 0.7224739789962769, "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/emo_Pain/f7a65885d0e107603bdce9d92b5fc3b68a68524db7b0b06f79b6538794e8bf5c.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": "f813ae4ca84e6fdd13206dd6a27bc008c94f53c80ac4409472eedfd717fc5290", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dime cuándo podemos hablar.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Tu mensaje llegó después de que me fuera.\n(Natural clear speech.) [3.8 seconds duration] Lo leí dos veces... y guardé el teléfono.\n(Natural clear speech.) [3.2 seconds duration] Dime cuándo podemos hablar.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "b60277e5b0e80ccb1cd3cffd673632b8cf9d95ca8b425c65e770a29edff6e591", "generation_seconds_batch": 22.361075052060187, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2410472631454468, "target_z": 1.7109375469570387, "ce": 5.871703147888184, "genuineness": 4.369274139404297, "identity": 0.6750400066375732, "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": 0.03999999999999915, "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/emo_Pain/f813ae4ca84e6fdd13206dd6a27bc008c94f53c80ac4409472eedfd717fc5290.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "f8141aca6266bac5990e11ae9e3472f8fa4785eb51b5dab2a736401a23fe79d1", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "afca5f602ce24b8c353953d58e4596055bcc2655fb087f6a008f3aa64cdd562c", "generation_seconds_batch": 22.54574990295805, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1328308880329132, "target_z": -0.21582030177274544, "ce": 6.046441555023193, "genuineness": 4.909013748168945, "identity": 0.6116495132446289, "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/emo_Pain/f8141aca6266bac5990e11ae9e3472f8fa4785eb51b5dab2a736401a23fe79d1.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-0", "key": "f83012e7fa16927b80eaaffc5c5e1a5332e24bbd21fc07aaea30a7495300876d", "bucket": "emo_Pain", "phase": "wording", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La puerta sigue abierta.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] He dejado tu abrigo sobre la silla.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "713487f013b2b588dd9010ebe3ef3c4b32067e1ee2fd31613851fb277fbda423", "generation_seconds_batch": 17.438399091013707, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07610009610652924, "target_z": -0.31445311487994854, "ce": 6.022613525390625, "genuineness": 4.413206577301025, "identity": 0.4449997544288635, "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/emo_Pain/f83012e7fa16927b80eaaffc5c5e1a5332e24bbd21fc07aaea30a7495300876d.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": "emo_Pain", "phase": "heldout", "candidate": "standard", "prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": "6baafada43b08ed7b2d708d5c670b4a56ff38f42db36d1ae68580cd182ec0569", "generation_seconds_batch": 19.595872953068465, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.0480155348777771, "target_z": -0.3632812589056453, "ce": 6.181467056274414, "genuineness": 4.545003414154053, "identity": 0.5094133615493774, "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/emo_Pain/f86feddc7af334e5c3d6691e715ece3168f8e7618bd3b63ca7d0171f412fdf9d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-de-0", "key": "f8f2100d3def58e96292ae7a73506b898a3898bd77ea9732c23d947edeeae331", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Die Tür ist noch offen.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] Die Tür ist noch offen.\n(Natural clear speech.) [3.8 seconds duration] Ich habe deinen Mantel auf den Stuhl gelegt.\n(Natural clear speech.) [3.2 seconds duration] Sag mir, wenn du bereit bist.", "seed": 1702, "scale": 0.0, "cfg": 4.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "9af21834b225d571f1386ad40ecb09e10f756d04eeae3eec6ad756f9479767ae", "generation_seconds_batch": 17.57663396105636, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.9804226160049438, "target_z": 1.2578125113344576, "ce": 5.71682071685791, "genuineness": 4.582659721374512, "identity": 0.5767114758491516, "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/emo_Pain/f8f2100d3def58e96292ae7a73506b898a3898bd77ea9732c23d947edeeae331.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "f9bde0a79a98b0afdbe0ab1269b3ebe59b218c793d67a838dfd3e42ab44c63ee", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Tell me when you are ready.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The door is still open.\n(Natural clear speech.) [3.8 seconds duration] I put your coat on the chair.\n(Natural clear speech.) [3.2 seconds duration] Tell me when you are ready.", "seed": 1702, "scale": 0.5, "cfg": 5.0, "ended": true, "frames_generated": 127, "seconds_generated": 10.16, "mp3_sha256": "4bd90253b90c377e14643d886d22a8ab253128bbceaa275544f3175f76413f62", "generation_seconds_batch": 22.43536167300772, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.0028902292251587, "target_z": 1.296874964377419, "ce": 6.213237762451172, "genuineness": 4.231201171875, "identity": 0.45602983236312866, "hypothesis": "The door is still open. I put your coat on the chair. Tell me when you are ready.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/manual-comparison-v1/audio/emo_Pain/f9bde0a79a98b0afdbe0ab1269b3ebe59b218c793d67a838dfd3e42ab44c63ee.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "fa3630f2b5fd0f5b57b56dffc14ae833356d750d9c3c0b46eeb60f35e7d54b29", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "e1643ee6b57b26c0f4969c72bb439165db41aa084422f6e961edd1aabe7c521a", "generation_seconds_batch": 17.04402014799416, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.1210353672504425, "target_z": -0.23632813107203088, "ce": 6.1894097328186035, "genuineness": 4.293961524963379, "identity": 0.7259973287582397, "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/emo_Pain/fa3630f2b5fd0f5b57b56dffc14ae833356d750d9c3c0b46eeb60f35e7d54b29.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "fa7afcf0299d2fed70d7a6efce1cdb8a4dd94c3a9f327afffac216b8d1d2b79e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "95e93bb258806c34798f48aae8887c873b615102e16b230f6401c544c662a448", "generation_seconds_batch": 24.066187088028528, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.059249356389045715, "target_z": -0.34375000647683296, "ce": 5.863759994506836, "genuineness": 4.65797233581543, "identity": 0.711528480052948, "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/emo_Pain/fa7afcf0299d2fed70d7a6efce1cdb8a4dd94c3a9f327afffac216b8d1d2b79e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "fabf7131161f77280c15af3245c7eb95a32056d1663cb7ae7d96b5006f9eba87", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "52c9991ec05bb6be303f53c5507301e4588ffeab4a87f4bc3ff65ddf778654bb", "generation_seconds_batch": 16.501453459030017, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.0305562019348145, "genuineness": 4.545003414154053, "identity": 0.7403742074966431, "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/emo_Pain/fabf7131161f77280c15af3245c7eb95a32056d1663cb7ae7d96b5006f9eba87.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "fb6fa7cfab86c10b466166d8bd79e20bfa57fd805d71980568921a24c813adfc", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "dacd72a284d99dfe3e0bcf31e5abf671e6132086335d64c79da87f37a09414ee", "generation_seconds_batch": 22.604677945026197, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.4501863121986389, "target_z": 0.33593752104970703, "ce": 5.863759994506836, "genuineness": 4.68307638168335, "identity": 0.7486299276351929, "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/emo_Pain/fb6fa7cfab86c10b466166d8bd79e20bfa57fd805d71980568921a24c813adfc.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "fbedc17d7742716ecc22ea1685a867c99dabb9bdcbee0966abc105679fefbe40", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "ab29e457515d0c8d58b9233a1287d5ce4a44088b19ad634fbaf69a987bc74d9a", "generation_seconds_batch": 18.77771246910561, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.0146708488464355, "genuineness": 4.68307638168335, "identity": 0.7335439324378967, "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/emo_Pain/fbedc17d7742716ecc22ea1685a867c99dabb9bdcbee0966abc105679fefbe40.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "fc359c8541bbfdb5e41460ba7fbf669e5ebdc9a8b09515ee9b782c96a25f1b46", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "acbcc24c082acb9f28875f62abc1cc6dc1b28047b402160b9ac4a01a4b17e051", "generation_seconds_batch": 22.114062309032306, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.340320587158203, "genuineness": 4.413206577301025, "identity": 0.4481359124183655, "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/emo_Pain/fc359c8541bbfdb5e41460ba7fbf669e5ebdc9a8b09515ee9b782c96a25f1b46.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-1", "key": "fc3d3581eb102359d1634933b1fb81adf37114db87830c39882f9e9f2ec6886a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] J’ai trouvé la petite boîte.\n(Natural clear speech.) [3.8 seconds duration] Elle était près de la vieille fenêtre.\n(Natural clear speech.) [3.2 seconds duration] Nous pouvons regarder ensemble ce qu’elle contient.", "seed": 1702, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "a54914e6f6422e9fc7573210a5d7e2592917860726033c5d2978d4ff1e53a699", "generation_seconds_batch": 22.363918137038127, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11878859996795654, "target_z": -0.2402343867392597, "ce": 5.744620323181152, "genuineness": 4.6956281661987305, "identity": 0.759441614151001, "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/emo_Pain/fc3d3581eb102359d1634933b1fb81adf37114db87830c39882f9e9f2ec6886a.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "fe9a75cc4d77a91f510dc950f853f57d59359677640af50a45497285998d0878", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "d4d0aad2bde4df1e1413ee5596de899e99a5888841964e2069c01384cb44afab", "generation_seconds_batch": 16.399933042936027, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.12889903783798218, "target_z": -0.22265626214406176, "ce": 6.2767791748046875, "genuineness": 4.7081804275512695, "identity": 0.5574000477790833, "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/emo_Pain/fe9a75cc4d77a91f510dc950f853f57d59359677640af50a45497285998d0878.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "feccf670d12a85cd4f879456d6c0e40d6a7ecd1825baf4d7c53085ac23b6e4c8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.5, "cfg": 2.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "5d0b444575cd1d39a1454525fa4025d04aa1ec9e525bf9ce5241a96c23f49daa", "generation_seconds_batch": 24.323153220932, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5029852390289307, "target_z": 0.4277343478782621, "ce": 5.772419452667236, "genuineness": 4.7081804275512695, "identity": 0.7480337619781494, "hypothesis": "La porte est encore ouverte. J'ai 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/manual-comparison-v1/audio/emo_Pain/feccf670d12a85cd4f879456d6c0e40d6a7ecd1825baf4d7c53085ac23b6e4c8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-es-1", "key": "fedcd7a05d6c5e7531a02ff02857ba731fbea0d42e1ab294e03c3bd3379ad11e", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] He encontrado la cajita.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Estaba junto a la ventana vieja.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "389041a32fcba2035d25bc53cc25723229a53575de13440394b0040917571048", "generation_seconds_batch": 23.50753080798313, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.8546037673950195, "target_z": 1.0390624012282976, "ce": 5.387199878692627, "genuineness": 4.319066047668457, "identity": 0.4666985273361206, "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/emo_Pain/fedcd7a05d6c5e7531a02ff02857ba731fbea0d42e1ab294e03c3bd3379ad11e.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-0", "key": "ff1654bb015ed3dc9274fbc73877e5be898d3c7d33dad78ba4c2e6563745f92d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The door is still open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I put your coat on the chair.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "8ba76bb8196fbaccec391dbd7530e56a15d016c703cdd775b5812be9f5099d39", "generation_seconds_batch": 24.084413825999945, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.11991198360919952, "target_z": -0.23828125890564528, "ce": 6.324435234069824, "genuineness": 4.394378662109375, "identity": 0.3860957622528076, "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/emo_Pain/ff1654bb015ed3dc9274fbc73877e5be898d3c7d33dad78ba4c2e6563745f92d.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": "ff43f4c6a55a3dc454feff25cfa6fd979843c8c085f1cf1005a2cd3b5d63ca17", "bucket": "emo_Pain", "phase": "heldout", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] The keys are on the kitchen table.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Take them with you when you leave.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] The keys are on the kitchen table.\n(Natural clear speech.) [3.8 seconds duration] I checked the drawer... then looked beside the window.\n(Natural clear speech.) [3.2 seconds duration] Take them with you when you leave.", "seed": 1702, "scale": 0.0, "cfg": 3.0, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "931328865b50a111c8fec327578e3f713fe0bbf6cd150546357329494db0b029", "generation_seconds_batch": 18.299639372969978, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.2507860064506531, "target_z": -0.010742164223881627, "ce": 6.046441555023193, "genuineness": 4.356721878051758, "identity": 0.47233444452285767, "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/emo_Pain/ff43f4c6a55a3dc454feff25cfa6fd979843c8c085f1cf1005a2cd3b5d63ca17.mp3" }, { "id": "emo_Pain-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": "emo_Pain-fr-0", "key": "ff47bfbd6c1da54ae8be1ea26d47fda4b218ae4cf74c0c61808afab82abd51d0", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] La porte est encore ouverte.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "neutral_prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects.\nSCRIPT:\n(Natural clear speech.) [3.2 seconds duration] La porte est encore ouverte.\n(Natural clear speech.) [3.8 seconds duration] J’ai posé ton manteau sur la chaise.\n(Natural clear speech.) [3.2 seconds duration] Dis-moi quand tu es prêt.", "seed": 1701, "scale": 0.75, "cfg": 2.5, "ended": true, "frames_generated": 128, "seconds_generated": 10.24, "mp3_sha256": "875f1b8bc64f13edd48e3439e49abcfde89f4e27b99fb49450f57a8fd2c94487", "generation_seconds_batch": 24.086277142050676, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 1.2859824895858765, "target_z": 1.7890624530429613, "ce": 5.816103935241699, "genuineness": 4.532451152801514, "identity": 0.5986429452896118, "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/emo_Pain/ff47bfbd6c1da54ae8be1ea26d47fda4b218ae4cf74c0c61808afab82abd51d0.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-0", "key": "ff5ea459a35cc6b167a0db67b83c53eb9fe433b4f84268020b044180c69b414d", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] De deur staat nog open.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Ik heb je jas op de stoel gelegd.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "aca08a3512a13f4638f1f49d4628a1b4ba56cafda9a8d22fa7e1ea03d907068a", "generation_seconds_batch": 22.821842935984023, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.49175143241882324, "target_z": 0.4082031213567815, "ce": 5.855817794799805, "genuineness": 4.250029563903809, "identity": 0.4428553581237793, "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/emo_Pain/ff5ea459a35cc6b167a0db67b83c53eb9fe433b4f84268020b044180c69b414d.mp3" }, { "id": "emo_Pain-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": "emo_Pain-nl-1", "key": "ffb30963e15ffb6d2fddb676bce2229da3463ab5a9ebc4f450f658f2f8ef72f8", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] Ik heb het kleine doosje gevonden.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] Het stond naast het oude raam.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "f11ae50239ad080005a4e0ca26bf6732ed6dc108e416fc06e2cedcff1eef27eb", "generation_seconds_batch": 21.792488071951084, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.5175892114639282, "target_z": 0.45312498380791766, "ce": 5.681078910827637, "genuineness": 4.18099308013916, "identity": 0.43590986728668213, "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/emo_Pain/ffb30963e15ffb6d2fddb676bce2229da3463ab5a9ebc4f450f658f2f8ef72f8.mp3" }, { "id": "emo_Pain-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": "emo_Pain-en-1", "key": "ffe4711f4fc34111b33740e66e64777bf11516d6191778b97952456a0a4e917a", "bucket": "emo_Pain", "phase": "grid", "candidate": "manual-abbf6e2776b7", "prompt": "GENERAL: One adult speaker, clean high-quality recording, no music or sound effects. The performance must convey a sense of desperate, near-breaking vulnerability. The pain is less about the injury and more about the inability to control the reaction to it.\nSCRIPT:\n(Begin with an extremely fragile, almost breathy onset, as if the speaker cannot summon the strength to project the initial sound.) [3.2 seconds duration] I found the little box.\n(Introduce brief, uncontrolled vocal fry or vocal breaks mid-sentence, punctuating the speech with involuntary physical reactions.) [3.8 seconds duration] It was beside the old window.\n(Conclude with a forced, strained projection on the final word, a momentary surge of desperate will overriding the physical distress.) [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": "adf371e2f3aa603c5de393eb9c0706be0d038238e38fb192852bfea09d8f32d9", "generation_seconds_batch": 18.47566448105499, "original_audio_reused": false, "scoring_contract": "singleton-EARS-ASR-Orange128-preMP3-v1", "metrics": { "valid_waveform": true, "target_raw": 0.13002242147922516, "target_z": -0.22070313431044736, "ce": 6.23706579208374, "genuineness": 4.632867813110352, "identity": 0.6524307727813721, "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/emo_Pain/ffe4711f4fc34111b33740e66e64777bf11516d6191778b97952456a0a4e917a.mp3" }, { "key": "original-emo_Pain-de-0-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "42c2d739b117c6cdb9a2fa6fdf987a8eb907d2507b30f0ec1d0adb48bd4b0fbe", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08171699941158295, "target_z": -0.3046875016192082, "ce": 6.173524379730225, "genuineness": 4.482243061065674, "identity": 0.5265872478485107, "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/emo_Pain/de/5b123061c2715734511dbf75.mp3" }, { "key": "original-emo_Pain-de-0-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "4ac4fae4322baee17fd326af9b96e25a44254705ab1d973cc7be4e671b32a010", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.181467056274414, "genuineness": 4.457139015197754, "identity": 0.7734271287918091, "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/emo_Pain/de/11fe990bcedb4fcd8e77124c.mp3" }, { "key": "original-emo_Pain-de-0-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "ba23e9773843625f898584d5fa52144bf775f99cbabbe7e0d117d13241f10957", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08845728635787964, "target_z": -0.29296876052485354, "ce": 6.1020402908325195, "genuineness": 4.457139015197754, "identity": 0.5986769795417786, "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/emo_Pain/de/987bf0c8e4b078a97aacb419.mp3" }, { "key": "original-emo_Pain-de-0-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "01e4f246b46c4b8f384e61a793c93b83ebf5e93ff4417ca44b96e52dc40bd8f0", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08733391761779785, "target_z": -0.29492186245113616, "ce": 6.109982967376709, "genuineness": 4.325342178344727, "identity": 0.6444048285484314, "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/emo_Pain/de/3fbe5634353f58e7d5d7e93d.mp3" }, { "key": "original-emo_Pain-de-0-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "701ba78a7abef8c279e3a93baab89a2a67d274bf026645dce5edd41543ca0356", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 6.260893821716309, "genuineness": 4.582659721374512, "identity": 0.43681272864341736, "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/emo_Pain/de/6588e7f6d559d15e82a60589.mp3" }, { "key": "original-emo_Pain-de-0-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "3c641fe556309c29110b16b22942bf5c4a428ed125a87a82eabdee7c77350e38", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 6.109982967376709, "genuineness": 4.519899368286133, "identity": 0.6650108695030212, "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/emo_Pain/de/14b7b67b7442a2cd9fe002a4.mp3" }, { "key": "original-emo_Pain-de-0-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "0145d204d891336f8b168f5c8d13256ffd84023b17984d20d0ab425e55625c01", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.268836498260498, "genuineness": 4.7081804275512695, "identity": 0.6022244691848755, "hypothesis": "Die Tür ist noch offen. Ich habe deinen Mantel auf den Stuhl gelegt. Sag mir, wenn du.", "wer": 0.10526315789473684, "duration_error_s": 0.040000000000000924, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Pain/de/067853493b668e363db97a9d.mp3" }, { "key": "original-emo_Pain-de-0-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "0b5b5fbdf75a9cda1a29805f95f82e89141259bb3bc47729bcf2a414b5d66c67", "task_id": "emo_Pain-de-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07385332882404327, "target_z": -0.31835937054717733, "ce": 6.260893821716309, "genuineness": 4.645420074462891, "identity": 0.49937522411346436, "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/emo_Pain/de/7c2d8bd07920e2c20bf05574.mp3" }, { "key": "original-emo_Pain-de-1-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "e8005a7a0e52b6333ccf675f405736da01744e38624229dc6999c961feca2607", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07272994518280029, "target_z": -0.3203124983807918, "ce": 6.38003396987915, "genuineness": 4.6705241203308105, "identity": 0.32390856742858887, "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/emo_Pain/de/f35e012715728c118d9599d2.mp3" }, { "key": "original-emo_Pain-de-1-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "10d56ac4aff008695a7762a51ec29f345d59fd8adc83e31cc6a1a7e64cba8130", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07497671246528625, "target_z": -0.31640624271356294, "ce": 6.332377910614014, "genuineness": 4.507347106933594, "identity": 0.5341755151748657, "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/emo_Pain/de/b8c7ea68e7950fcad0fade30.mp3" }, { "key": "original-emo_Pain-de-1-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "00a3440e059e9a41e2f3402d97666dfce64a2c9f35cb877156836f68f1326697", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.036781713366508484, "target_z": -0.38281251133445765, "ce": 6.443574905395508, "genuineness": 4.369274139404297, "identity": 0.6112103462219238, "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/emo_Pain/de/78b801fa7f9994e7d216efe7.mp3" }, { "key": "original-emo_Pain-de-1-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "5e26ca53c616c86a911c15363b9f8ac3b8167dcaf2b2677b5822255f73c31832", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.04239863157272339, "target_z": -0.3730468721663856, "ce": 6.332377910614014, "genuineness": 4.494794845581055, "identity": 0.5683363676071167, "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/emo_Pain/de/f457cabe578def1b12957817.mp3" }, { "key": "original-emo_Pain-de-1-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "0717029e9ff05baef581bb90c6bc012d17fee4ecaf9466599ea1b611ea11fab9", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.058125972747802734, "target_z": -0.34570313431044736, "ce": 6.2767791748046875, "genuineness": 4.6705241203308105, "identity": 0.5755237340927124, "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/emo_Pain/de/efd77b1fb8f2e985ecd5dc31.mp3" }, { "key": "original-emo_Pain-de-1-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "54fbac8c306a047fcc9a3bb8df514cb7fc187ba78735b8c34646612d5b85e0ed", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.356205940246582, "genuineness": 4.65797233581543, "identity": 0.691216766834259, "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/emo_Pain/de/e22b2ca764b50e745d06e34c.mp3" }, { "key": "original-emo_Pain-de-1-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "d4194d902f13acd8773e12a8a1aafe33f3d49a23c04c702cd60101bb2bd5c3d4", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.054755836725234985, "target_z": -0.35156249190395883, "ce": 6.316492557525635, "genuineness": 4.419482707977295, "identity": 0.6736986041069031, "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/emo_Pain/de/438bfa2ce6476876402c77a6.mp3" }, { "key": "original-emo_Pain-de-1-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Wir können gemeinsam hineinschauen.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "d3061de8026e836eeab69d262ce4c7a9f9c500876a3ba516ae65f284c94e87fe", "task_id": "emo_Pain-de-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.0480155348777771, "target_z": -0.3632812589056453, "ce": 6.3641486167907715, "genuineness": 4.595211505889893, "identity": 0.5738694667816162, "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/emo_Pain/de/112549d73f25606e8692b762.mp3" }, { "key": "original-emo_Pain-en-0-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "7f79214c98130997a6814cf4bf69b3800bc501223728eee06a4f8a0c47b513a7", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08621053397655487, "target_z": -0.2968749902847506, "ce": 6.3641486167907715, "genuineness": 4.193545341491699, "identity": 0.40654778480529785, "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/emo_Pain/en/4c7ebe15be35e614ab6c93b3.mp3" }, { "key": "original-emo_Pain-en-0-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "60822e22c58b2849db4725087b73286f673619be280bd58f2d36102f58af7af4", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08733391761779785, "target_z": -0.29492186245113616, "ce": 6.24500846862793, "genuineness": 4.105680465698242, "identity": 0.35830044746398926, "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/emo_Pain/en/7220ffc758fdb343b89d17e9.mp3" }, { "key": "original-emo_Pain-en-0-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "4fb82d030a8834aa324202f59ffbf97242f579a06deee40c8ee5d4b82751c5fe", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.11822691559791565, "target_z": -0.24121093770240104, "ce": 6.24500846862793, "genuineness": 4.432034492492676, "identity": 0.35065579414367676, "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/emo_Pain/en/cc7fd6bede3d41081af80998.mp3" }, { "key": "original-emo_Pain-en-0-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "8f2b257b0426ad03b73e65410160c630a2c2b4022b36fcb42bba324ece9319f2", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.181467056274414, "genuineness": 3.886019229888916, "identity": 0.4231913685798645, "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/emo_Pain/en/186c37a4219255a192d43332.mp3" }, { "key": "original-emo_Pain-en-0-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "186514c7ae45430452dbd02fa4fa56440a210ee17d9b6f4b1c4ede1a396ee4ad", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09182743728160858, "target_z": -0.2871093770240103, "ce": 6.078212261199951, "genuineness": 3.9676074981689453, "identity": 0.3033413887023926, "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/emo_Pain/en/5f8de6e4f5c9dcbf479a7975.mp3" }, { "key": "original-emo_Pain-en-0-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "ae172c4565780d39524fb2f9e6068e212c4b987d4b0eb2c1fb4e8bdc400bdbc6", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07497671246528625, "target_z": -0.31640624271356294, "ce": 6.2767791748046875, "genuineness": 4.080576419830322, "identity": 0.19997310638427734, "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/emo_Pain/en/09e520b0f6b807a88aa397c8.mp3" }, { "key": "original-emo_Pain-en-0-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "3f6ddb8d1e06fe6b55b4bc46e296e127ea84edf1c6dc2659c1cb77ddb1b5e54b", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09856772422790527, "target_z": -0.27539063592965557, "ce": 6.173524379730225, "genuineness": 4.042920112609863, "identity": 0.361219197511673, "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/emo_Pain/en/d901486722713a3579fc4633.mp3" }, { "key": "original-emo_Pain-en-0-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "ba0d02093b37117938d41a33b5c28032a58ede9e21cf0597677abae44ac5ee4c", "task_id": "emo_Pain-en-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08733391761779785, "target_z": -0.29492186245113616, "ce": 6.197352409362793, "genuineness": 4.193545341491699, "identity": 0.2608886957168579, "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/emo_Pain/en/188f68cb954086dad21eec1c.mp3" }, { "key": "original-emo_Pain-en-1-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "4656116ddf01bfb2d3d6153291452904c17dcfe9e01c300ccbe27970b0a22929", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08284038305282593, "target_z": -0.3027343737855938, "ce": 6.348263263702393, "genuineness": 4.65797233581543, "identity": 0.32861462235450745, "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/emo_Pain/en/c6762b8aed9872a96713efc4.mp3" }, { "key": "original-emo_Pain-en-1-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "ba2f66bc02a1d6ea644111323fd769e1e7e8409d857837a6c26d7cd95a1dcf2e", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.10418464243412018, "target_z": -0.2656249967615835, "ce": 6.356205940246582, "genuineness": 4.720732688903809, "identity": 0.6141794919967651, "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/emo_Pain/en/6959b4a7351eab0b4fb6a8e5.mp3" }, { "key": "original-emo_Pain-en-1-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "20981f3f52fb80d56ce8cdf0f67cf80601752619a192cdb0e5f26cbb61f05d3f", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.12328213453292847, "target_z": -0.23242187540480205, "ce": 6.4515180587768555, "genuineness": 4.68307638168335, "identity": 0.439589262008667, "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/emo_Pain/en/46ff84aa929ee8e117d5487e.mp3" }, { "key": "original-emo_Pain-en-1-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "1a54f8b4bc7b53745a0e28c8dd4b5fc61a1ec897e0351d73a9d802353ac0bcfe", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.10081449151039124, "target_z": -0.27148438026242677, "ce": 6.157639026641846, "genuineness": 4.432034492492676, "identity": 0.34934520721435547, "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/emo_Pain/en/7965593160db5d485d18dba3.mp3" }, { "key": "original-emo_Pain-en-1-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "de2ad6f0911cd730cd0b4ff98befbd3f87eb46eb7b5cec886eb8db1e4cf9ed67", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.10980154573917389, "target_z": -0.25585938350084325, "ce": 6.332377910614014, "genuineness": 4.557555198669434, "identity": 0.25155386328697205, "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/emo_Pain/en/cf51a2595d7bc434d5a93390.mp3" }, { "key": "original-emo_Pain-en-1-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "7994c8c6b936b84c1ebf87a5985af59de5853b50fb3faa58e8f8b3478bdc6ac7", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06935979425907135, "target_z": -0.326171881881635, "ce": 6.41180419921875, "genuineness": 4.68307638168335, "identity": 0.4807351231575012, "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/emo_Pain/en/0bd6686ecaebbe88f8c42836.mp3" }, { "key": "original-emo_Pain-en-1-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "327c1d5bffd4ad15b18fb3b706dfc76128ee48a172aa05a219af9086d8694982", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.08059361577033997, "target_z": -0.30664062945282267, "ce": 6.24500846862793, "genuineness": 4.457139015197754, "identity": 0.3310950994491577, "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/emo_Pain/en/bab19f83c9f5534686c46f60.mp3" }, { "key": "original-emo_Pain-en-1-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "1878f0a172749e941fdf7ddf06822978d72aeaebb16b484ef0817883f8c10658", "task_id": "emo_Pain-en-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09519758820533752, "target_z": -0.28124999352316704, "ce": 6.308549880981445, "genuineness": 4.65797233581543, "identity": 0.32265251874923706, "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/emo_Pain/en/4163b6f49871798806841b93.mp3" }, { "key": "original-emo_Pain-es-0-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "df386d3908204db70c66a19020b92a81b7514e4f01a7e56a2e17fe394cf80dd4", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 5.903473377227783, "genuineness": 4.595211505889893, "identity": 0.6032326221466064, "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/emo_Pain/es/8725ce9fc9aad302cb61ecb5.mp3" }, { "key": "original-emo_Pain-es-0-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "3da5af8050743e689ee0b2532cfe32ce418b13a60730f7a4b5fc01a4b86a3209", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.04689215123653412, "target_z": -0.3652343867392597, "ce": 5.9431867599487305, "genuineness": 3.8922953605651855, "identity": 0.6279011964797974, "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/emo_Pain/es/b1e24e393ed7ef36f9d4946c.mp3" }, { "key": "original-emo_Pain-es-0-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "3a6e78e8bc9c1d3de5c1e0941c3c9f2bdc7a144c45a9b70c1125bf9b6695ff7e", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.1030612587928772, "target_z": -0.2675781245951979, "ce": 6.157639026641846, "genuineness": 4.381826400756836, "identity": 0.7054051160812378, "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/emo_Pain/es/37f330756eeae668c571f755.mp3" }, { "key": "original-emo_Pain-es-0-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "9f4f87d36598c801d4418d3c4f2da1ee50c3a52f5cc62bbc199c0f2b0044c6ce", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.04689215123653412, "target_z": -0.3652343867392597, "ce": 5.935244083404541, "genuineness": 4.005263805389404, "identity": 0.5513920783996582, "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/emo_Pain/es/3f0b2801fe163872f156c561.mp3" }, { "key": "original-emo_Pain-es-0-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "47f80c87f4db138d7c0f41b1d17bda9d543fd97e90989272204d7802d14c77ca", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 5.919358730316162, "genuineness": 4.325342178344727, "identity": 0.5823712348937988, "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/emo_Pain/es/8a96cf55a37aed878973c128.mp3" }, { "key": "original-emo_Pain-es-0-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "c7200db422aa0b8bc3e8936a8b7ce0d6e8de64c935892d7ad30fca6f52663de1", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.04576878249645233, "target_z": -0.36718748866554235, "ce": 5.796247482299805, "genuineness": 3.9173994064331055, "identity": 0.6575822830200195, "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/emo_Pain/es/ced09c172182de3937b7c10f.mp3" }, { "key": "original-emo_Pain-es-0-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "21ce88cd96a10bff662124103890763d545440f05350384fd0578c23f4ddd176", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.058125972747802734, "target_z": -0.34570313431044736, "ce": 6.046441555023193, "genuineness": 4.118232727050781, "identity": 0.572709321975708, "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/emo_Pain/es/68bf2b4774406254549fdb6a.mp3" }, { "key": "original-emo_Pain-es-0-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] Dime cuándo estás listo.", "directions": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "a79d7fa85c4d052a7f1f2a9b7f92635aa5b86b021c16fa52a6783de7a65f0ace", "task_id": "emo_Pain-es-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] La puerta sigue abierta.\n(Intense Pain. 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(Intense Pain. 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": 0.036781713366508484, "target_z": -0.38281251133445765, "ce": 6.078212261199951, "genuineness": 3.8420867919921875, "identity": 0.5014321208000183, "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/emo_Pain/es/8522fa0a27241dd8402f32a5.mp3" }, { "key": "original-emo_Pain-es-1-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "a55cafcc86babd5150aab6eb71d43a53f197f71bc4adb0ff14ba531fbd456bbd", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.06149612367153168, "target_z": -0.3398437508096041, "ce": 6.046441555023193, "genuineness": 4.557555198669434, "identity": 0.7710758447647095, "hypothesis": "He encontrado consignado en la cajita. Estaba junto a la ventana vieja. Podemos mirar juntos lo que hay dentro.", "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/audio/emo_Pain/es/8032f17fa64d6f949bab1902.mp3" }, { "key": "original-emo_Pain-es-1-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "eee56c44e3ce0dd69e39a77ad6436d58ac33facf777ffe4222d9d3db5b630d3c", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.356205940246582, "genuineness": 4.796044826507568, "identity": 0.7051689624786377, "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/emo_Pain/es/a97cc364c59006372b00fe9e.mp3" }, { "key": "original-emo_Pain-es-1-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "7b97681176e2fd2a0634685450536bf0f555d94c69eb23719bfdbc8fc57ffc74", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.030041426420211792, "target_z": -0.3945312524288124, "ce": 6.181467056274414, "genuineness": 4.482243061065674, "identity": 0.5471856594085693, "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/emo_Pain/es/4f8febf6274804ffc75a047c.mp3" }, { "key": "original-emo_Pain-es-1-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "8233da047b0faddf31decd9f5cf4b4993679789c7e94c9496607e41a4271bed5", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.07048317790031433, "target_z": -0.3242187540480206, "ce": 6.133810997009277, "genuineness": 4.632867813110352, "identity": 0.7898926734924316, "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/emo_Pain/es/d8fea0ba9d73ab0edb3a5568.mp3" }, { "key": "original-emo_Pain-es-1-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "27bca0e186a4e31c9ed402b01635636fea4dfafc44f1f6b2ee7b5a257a30622b", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.034534960985183716, "target_z": -0.3867187410943547, "ce": 6.062326908111572, "genuineness": 4.494794845581055, "identity": 0.6704084873199463, "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/audio/emo_Pain/es/371bdde0ab75fbe3e9b231cc.mp3" }, { "key": "original-emo_Pain-es-1-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "fe0b798110af83a21f49a7ef40a9378f1c3731b52fa2d4b7bde976fba7e55dd4", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.04689215123653412, "target_z": -0.3652343867392597, "ce": 6.046441555023193, "genuineness": 4.7081804275512695, "identity": 0.569732666015625, "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/emo_Pain/es/b0e1994e00977e9b96a726a2.mp3" }, { "key": "original-emo_Pain-es-1-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "03f36087e58bff9bbed57b83428a8eb8f7d5246434f83767477f0c6b9ade5989", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.05250906944274902, "target_z": -0.3554687475711876, "ce": 6.038498878479004, "genuineness": 4.545003414154053, "identity": 0.6535241603851318, "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/emo_Pain/es/1f11cfa7f1b4b6c119e878a1.mp3" }, { "key": "original-emo_Pain-es-1-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "a3d62cdc7db67fb31df7681f31429b529182ee5b1938ed337ec91b9d0b23a447", "task_id": "emo_Pain-es-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.) [3.2 seconds duration] He encontrado la cajita.\n(Intense Pain. 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(Intense Pain. 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": 0.030041426420211792, "target_z": -0.3945312524288124, "ce": 6.054384231567383, "genuineness": 4.432034492492676, "identity": 0.6184645891189575, "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/emo_Pain/es/7d570b3aeddbb6e12a79c9e9.mp3" }, { "key": "original-emo_Pain-fr-0-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "d4cec5f36a4992c1b458d2fc35558abc618bb74f1518b3352d7b284aec3516bd", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.05138568580150604, "target_z": -0.3574218754048021, "ce": 6.149696350097656, "genuineness": 4.432034492492676, "identity": 0.7888714075088501, "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/emo_Pain/fr/95c5d0b344f1df8f1f38e5c1.mp3" }, { "key": "original-emo_Pain-fr-0-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "4b49cd8b6c0f9ed9fe51f74d7c850a1e89562830305c06e035cc76e4e54d0a9e", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 5.879645347595215, "genuineness": 4.2751336097717285, "identity": 0.6529605388641357, "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/emo_Pain/fr/6d70d8786fc7430f3a461769.mp3" }, { "key": "original-emo_Pain-fr-0-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "b4bfa224cdda431a105b9fedf5643aeaaa65dae5fe723ab6c92c00becd901b45", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06486627459526062, "target_z": -0.3339843673087609, "ce": 6.141753673553467, "genuineness": 4.65797233581543, "identity": 0.7941343188285828, "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/emo_Pain/fr/ca3b97eaadda3b9fe80ccacb.mp3" }, { "key": "original-emo_Pain-fr-0-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "4d6ea138dd33b96cb8f5aa8d5b1231ac3202cd36d16e9a051b82bbf87db99a12", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.055879220366477966, "target_z": -0.34960936407034443, "ce": 6.070269584655762, "genuineness": 4.582659721374512, "identity": 0.647834837436676, "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/emo_Pain/fr/67321968d6f2da20271fadba.mp3" }, { "key": "original-emo_Pain-fr-0-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "f00b15f97cce6741350bff39126124673f8ea6b7324ce4af4e993ff06d128255", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.02442452311515808, "target_z": -0.40429686568955264, "ce": 6.062326908111572, "genuineness": 4.507347106933594, "identity": 0.6606736183166504, "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/emo_Pain/fr/b27288dffb7dc2d55002c61c.mp3" }, { "key": "original-emo_Pain-fr-0-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "76febf5e3c8bb72352b8c0eb3d04a37156d91f921861ee7f8590652b141b73a0", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.02891804277896881, "target_z": -0.39648438026242677, "ce": 6.086154937744141, "genuineness": 4.482243061065674, "identity": 0.7853033542633057, "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/emo_Pain/fr/d49a2d45a67c5b0e0d3bd999.mp3" }, { "key": "original-emo_Pain-fr-0-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "e1e5a23076060ffb4b1d5a836ef5dc46ec8695e7b03a525e258374f830d2db6e", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.04576878249645233, "target_z": -0.36718748866554235, "ce": 6.117925643920898, "genuineness": 4.419482707977295, "identity": 0.7744425535202026, "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/emo_Pain/fr/2d0b9c1d9f1756d1e88f6a11.mp3" }, { "key": "original-emo_Pain-fr-0-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "24b8a274a73c76df6c6ce7d1eb1f5d8cdb7dade3937983ba4514c9a100be1e8e", "task_id": "emo_Pain-fr-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06261950731277466, "target_z": -0.3378906229759897, "ce": 6.046441555023193, "genuineness": 4.080576419830322, "identity": 0.7484369277954102, "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/emo_Pain/fr/218474078f0edb97dbb9d4a4.mp3" }, { "key": "original-emo_Pain-fr-1-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "ce8945d346bacdcbb8aeda714f44c5aefb2584880ce5519c09781ce0be87c1a4", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06935979425907135, "target_z": -0.326171881881635, "ce": 6.141753673553467, "genuineness": 4.444586753845215, "identity": 0.6944689154624939, "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/emo_Pain/fr/0f6dde38ceb62f485956b073.mp3" }, { "key": "original-emo_Pain-fr-1-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "2adfff41cb2d0e4a95756deb0ba79ea5df8cb4fd37424ce413d6e4ad27ec63b7", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07722347974777222, "target_z": -0.31249998704633414, "ce": 6.078212261199951, "genuineness": 4.557555198669434, "identity": 0.8103022575378418, "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/emo_Pain/fr/5b873611f0ac9590fbb2dcf9.mp3" }, { "key": "original-emo_Pain-fr-1-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "ef6725bb770ac4f579af455e20046331e70e964427989a721c19e98c740c7370", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.10980154573917389, "target_z": -0.25585938350084325, "ce": 6.260893821716309, "genuineness": 4.758388519287109, "identity": 0.7125985622406006, "hypothesis": "J'ai trouvé la petite boîte. Elle était hai 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/audio/emo_Pain/fr/f5755712ee51f8222461d0c7.mp3" }, { "key": "original-emo_Pain-fr-1-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "bfc683405b4ec9947dd985a043bbb2fc9bbf6872d61fb3cfce191232036b16da", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.078212261199951, "genuineness": 4.595211505889893, "identity": 0.7456585168838501, "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/emo_Pain/fr/2da987d53f2822095e6588b3.mp3" }, { "key": "original-emo_Pain-fr-1-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "3566702c339be7dd0ad6596801d700e77c7bb954749ced84b038a9dc2c4d5967", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.058125972747802734, "target_z": -0.34570313431044736, "ce": 6.141753673553467, "genuineness": 4.319066047668457, "identity": 0.5892872214317322, "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/emo_Pain/fr/bcc94bb9635c65c16c32d4e2.mp3" }, { "key": "original-emo_Pain-fr-1-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "3672a387b17c652d197094bcb8e470402d388f344bf3659c77eb4a8743ee3b1b", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.086154937744141, "genuineness": 4.532451152801514, "identity": 0.7471016645431519, "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/emo_Pain/fr/8a653202b860623dcf9e8129.mp3" }, { "key": "original-emo_Pain-fr-1-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "c8afc8f4b85910fc612b4c9df811987081cfeed48a50ae80038fe06402f804c8", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07385332882404327, "target_z": -0.31835937054717733, "ce": 6.157639026641846, "genuineness": 4.507347106933594, "identity": 0.7174867987632751, "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/emo_Pain/fr/24891cc6815833a8fef633b3.mp3" }, { "key": "original-emo_Pain-fr-1-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "e3332d2c229d76bea2873154f3606a6760f30d0b4426f79acee8d11a32a32b6a", "task_id": "emo_Pain-fr-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06374289095401764, "target_z": -0.3359374951423753, "ce": 6.117925643920898, "genuineness": 4.545003414154053, "identity": 0.7639805674552917, "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/emo_Pain/fr/7e4a9fc20f6f927018008e5b.mp3" }, { "key": "original-emo_Pain-nl-0-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "9e4bbe3a54e5c9cd653c8a85872cf9b46f0050b21d11f43d1a308a960b42d7e7", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.031164810061454773, "target_z": -0.3925781245951979, "ce": 6.149696350097656, "genuineness": 4.362998008728027, "identity": 0.7462737560272217, "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/emo_Pain/nl/76931ac15696220bdf54047d.mp3" }, { "key": "original-emo_Pain-nl-0-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "59f1bb95600e661a564210dccf423a2a97d4b14d3007ad38c04abfad11a2c5e7", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.04576878249645233, "target_z": -0.36718748866554235, "ce": 6.173524379730225, "genuineness": 4.174716949462891, "identity": 0.5398041009902954, "hypothesis": "De deur staat nog open. Ik heb je jas op de stoel gelegd. Zeg maar wanneer je klaar bent.", "wer": 0.0, "duration_error_s": 0.03999999999999915, "model_predictions_not_human_ratings": true }, "audio_url": "https://huggingface.co/datasets/laion/Humaneness-Voice-Large-1.5-Intense-Bucket-LoRAs-Demo-Audio/resolve/main/audio/emo_Pain/nl/e888ea772c62eafe94ec35b9.mp3" }, { "key": "original-emo_Pain-nl-0-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "b53cf526596460d3bd686da6139398706471d4791e51a3ebb7186e771062abb9", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.06823641061782837, "target_z": -0.3281250097152494, "ce": 6.006728172302246, "genuineness": 4.532451152801514, "identity": 0.5080846548080444, "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/emo_Pain/nl/47432225f857c31536a96d23.mp3" }, { "key": "original-emo_Pain-nl-0-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "fa6444f7ced9b8cc86a659ca147cf04af5db100719a2dcd617a4c5951acf5ec7", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.04015186429023743, "target_z": -0.3769531278336144, "ce": 6.078212261199951, "genuineness": 4.080576419830322, "identity": 0.5808665752410889, "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/emo_Pain/nl/74536d5912af24f900de4283.mp3" }, { "key": "original-emo_Pain-nl-0-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "d372ecda844f418a21889b569d04b4bc4f8b8f1f879d72c86ec62363f24f4919", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.030041426420211792, "target_z": -0.3945312524288124, "ce": 6.173524379730225, "genuineness": 4.319066047668457, "identity": 0.7323188781738281, "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/emo_Pain/nl/4a2db3b09f46151339e7336a.mp3" }, { "key": "original-emo_Pain-nl-0-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "72a16fdb01b9897400c7cdd5c5511b836583ef6866a8fef0695c7f4ee4bc924b", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.09295082092285156, "target_z": -0.2851562491903959, "ce": 6.133810997009277, "genuineness": 4.369274139404297, "identity": 0.6941431164741516, "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/emo_Pain/nl/e632e29062a67a13f255962e.mp3" }, { "key": "original-emo_Pain-nl-0-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "7bddab9e2c2a61e1bcee0f04d085fddaba701adf6fba31860971fe3268319754", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.060372740030288696, "target_z": -0.3417968786432185, "ce": 6.125868320465088, "genuineness": 4.3881025314331055, "identity": 0.6654973030090332, "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/emo_Pain/nl/df51e0db4cb88e75865dc5e1.mp3" }, { "key": "original-emo_Pain-nl-0-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "06b971fb6736513f6fe020d9dabaa2e94a2a674cb10c12c14f45f7e96b3c5be6", "task_id": "emo_Pain-nl-0", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.032288193702697754, "target_z": -0.3906249967615835, "ce": 6.141753673553467, "genuineness": 4.268857479095459, "identity": 0.6754868030548096, "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/emo_Pain/nl/105b11b1d88df66357f97ade.mp3" }, { "key": "original-emo_Pain-nl-1-seed1701-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.0, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "fff29fc76c622241bd7b1738dda54e2b0e032792e8f33b0c5d75b771346b677f", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.039028480648994446, "target_z": -0.3789062556672288, "ce": 6.229123115539551, "genuineness": 4.494794845581055, "identity": 0.6387994885444641, "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/emo_Pain/nl/9eeb0542ff04f050ee41e949.mp3" }, { "key": "original-emo_Pain-nl-1-seed1701-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word." ], "seconds": 10.2, "frames": 128, "seeds": [ 1701, 1702 ], "scales": [ 0.0, 0.5, 0.75, 1.0 ], "cfg": 1.0, "reference": true, "case_type": "Neutral wording, target emotion/style varies only through acoustic performance", "seed": 1701, "scale": 0.5, "frames_generated": 127, "seconds_generated": 10.16, "ended": true, "contract_sha256": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "b649f6530e92919125bf559a3df33a3449005507f67afee0fabd050c84393d1e", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.07497671246528625, "target_z": -0.31640624271356294, "ce": 6.1894097328186035, "genuineness": 4.394378662109375, "identity": 0.6479153633117676, "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/emo_Pain/nl/c76d76e8578255dd8c49a755.mp3" }, { "key": "original-emo_Pain-nl-1-seed1701-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "832973d3e34871a9e1af780b8c36d8be188b2e093538c1596ec414bade2dca64", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.02779465913772583, "target_z": -0.39843750809604117, "ce": 5.990842819213867, "genuineness": 4.306513786315918, "identity": 0.5925227999687195, "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/emo_Pain/nl/f96a69848b07ec539ec65424.mp3" }, { "key": "original-emo_Pain-nl-1-seed1701-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "557f8c0f6bc1ef2862852c25f5e6070f3c8b7709cc434cb4da67e38e46245ff2", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.0659896582365036, "target_z": -0.33203123947514646, "ce": 6.205295085906982, "genuineness": 4.406930446624756, "identity": 0.6342769265174866, "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/emo_Pain/nl/67c2bafdad680dac7c253034.mp3" }, { "key": "original-emo_Pain-nl-1-seed1702-scale0.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "de006ffac5dafa862d7401fc9cf7353526e79b5b0dd045093db3fefce6623496", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.04352201521396637, "target_z": -0.3710937443327712, "ce": 6.197352409362793, "genuineness": 4.482243061065674, "identity": 0.7272898554801941, "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/emo_Pain/nl/24a5b00949daa4bcc36a5615.mp3" }, { "key": "original-emo_Pain-nl-1-seed1702-scale0.5", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "0ce0c5a8c23e875247244806b7f3ddc9a939b671e595221ae3df88b5bcd888d4", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.030041426420211792, "target_z": -0.3945312524288124, "ce": 5.935244083404541, "genuineness": 4.381826400756836, "identity": 0.6082567572593689, "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/emo_Pain/nl/b41c20d3864ed0bd978b5d2d.mp3" }, { "key": "original-emo_Pain-nl-1-seed1702-scale0.75", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "4933826b829a4ff0dda1e8848e45ec4623f90c2351f88900b2ebf41295c116a4", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.036781713366508484, "target_z": -0.38281251133445765, "ce": 6.1020402908325195, "genuineness": 4.413206577301025, "identity": 0.4933948814868927, "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/emo_Pain/nl/5ece18e68eeaeb8583e149ea.mp3" }, { "key": "original-emo_Pain-nl-1-seed1702-scale1.0", "bucket": "emo_Pain", "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. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": [ "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.", "Intense Pain. 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": "e0bd8fabf5754f77ee4a43bc3edc238869ca55913c048937a5dda2d85e1f994f", "mp3_sha256": "df9aa4b5ac6a135a80ecbce93ae3d32d2533aa7cac2da2c66681a5e16f7edf98", "task_id": "emo_Pain-nl-1", "phase": "standard", "candidate": "standard", "neutral_prompt": "GENERAL: One adult speaker in a clean recording. Intense Pain. Make the requested delivery clearly audible, not merely implied by the words. Preserve every word.\nSCRIPT:\n(Intense Pain. 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(Intense Pain. 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(Intense Pain. 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": 0.036781713366508484, "target_z": -0.38281251133445765, "ce": 6.205295085906982, "genuineness": 4.369274139404297, "identity": 0.44410306215286255, "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/emo_Pain/nl/242456bea22039c01af1d05c.mp3" } ], "source_standard_audio_unchanged": true }